Assignment 1 (6).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "1b21afb2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from dataprep.eda import create_report\n",
    "import sweetviz as sv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "108b73df",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import pytz\n",
    "import numpy as np\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "20188b99",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from torchtext.data.utils import get_tokenizer\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import torch\n",
    "from torch import nn \n",
    "import torch.nn.functional as F\n",
    "from torch import utils\n",
    "\n",
    "torch.manual_seed(0)\n",
    "np.random.seed(0)\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "aae9812b",
   "metadata": {},
   "outputs": [],
   "source": [
    "train=pd.read_csv('https://raw.githubusercontent.com/zerickzhan/Group-20/main/review_train.csv')\n",
    "challenge=pd.read_csv('https://raw.githubusercontent.com/zerickzhan/Group-20/main/review_challenge.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "809a421a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>570306133677760513</td>\n",
       "      <td>cairdin</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>What said.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 11:35</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>570301130888122368</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3486</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>plus you've added commercials to the experienc...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 11:15</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>570301083672813571</td>\n",
       "      <td>yvonnalynn</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6837</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>I didn't today... Must mean I need to take ano...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 11:15</td>\n",
       "      <td>Lets Play</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>570301031407624196</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.7033</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>it's really aggressive to blast obnoxious \"ent...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 11:15</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>570300817074462722</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>and it's a really big bad thing about it</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 11:14</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13757</th>\n",
       "      <td>569587686496825344</td>\n",
       "      <td>KristenReenders</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3487</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>thank you we got on a different flight to Chic...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 12:01</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13758</th>\n",
       "      <td>569587371693355008</td>\n",
       "      <td>itsropes</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>leaving over minutes Late Flight. No warnings ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 11:59</td>\n",
       "      <td>Texas</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13759</th>\n",
       "      <td>569587242672398336</td>\n",
       "      <td>sanyabun</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>Please bring American Airlines to #BlackBerry</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 11:59</td>\n",
       "      <td>Nigeria,lagos</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13760</th>\n",
       "      <td>569587188687634433</td>\n",
       "      <td>SraJackson</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6659</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>you have my money, you change my flight, and d...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 11:59</td>\n",
       "      <td>New Jersey</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13761</th>\n",
       "      <td>569587140490866689</td>\n",
       "      <td>daviddtwu</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6771</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>we have ppl so we need know how many seats are...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 11:58</td>\n",
       "      <td>dallas, TX</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>13762 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id        user_name airline_sentiment  \\\n",
       "0      570306133677760513          cairdin           neutral   \n",
       "1      570301130888122368         jnardino          positive   \n",
       "2      570301083672813571       yvonnalynn           neutral   \n",
       "3      570301031407624196         jnardino          negative   \n",
       "4      570300817074462722         jnardino          negative   \n",
       "...                   ...              ...               ...   \n",
       "13757  569587686496825344  KristenReenders          positive   \n",
       "13758  569587371693355008         itsropes          negative   \n",
       "13759  569587242672398336         sanyabun           neutral   \n",
       "13760  569587188687634433       SraJackson          negative   \n",
       "13761  569587140490866689        daviddtwu           neutral   \n",
       "\n",
       "       sentiment_confidence         negative_reason  \\\n",
       "0                    1.0000                     NaN   \n",
       "1                    0.3486                     NaN   \n",
       "2                    0.6837                     NaN   \n",
       "3                    1.0000              Bad Flight   \n",
       "4                    1.0000              Can't Tell   \n",
       "...                     ...                     ...   \n",
       "13757                0.3487                     NaN   \n",
       "13758                1.0000  Customer Service Issue   \n",
       "13759                1.0000                     NaN   \n",
       "13760                1.0000  Customer Service Issue   \n",
       "13761                0.6771                     NaN   \n",
       "\n",
       "       negative_reason_confidence airline_name  \\\n",
       "0                             NaN     EgyptAir   \n",
       "1                          0.0000     EgyptAir   \n",
       "2                             NaN     EgyptAir   \n",
       "3                          0.7033     EgyptAir   \n",
       "4                          1.0000     EgyptAir   \n",
       "...                           ...          ...   \n",
       "13757                      0.0000       Qantas   \n",
       "13758                      1.0000       Qantas   \n",
       "13759                         NaN       Qantas   \n",
       "13760                      0.6659       Qantas   \n",
       "13761                      0.0000       Qantas   \n",
       "\n",
       "                                             review_text  thumbup_count  \\\n",
       "0                                             What said.              0   \n",
       "1      plus you've added commercials to the experienc...              0   \n",
       "2      I didn't today... Must mean I need to take ano...              0   \n",
       "3      it's really aggressive to blast obnoxious \"ent...              0   \n",
       "4               and it's a really big bad thing about it              0   \n",
       "...                                                  ...            ...   \n",
       "13757  thank you we got on a different flight to Chic...              0   \n",
       "13758  leaving over minutes Late Flight. No warnings ...              0   \n",
       "13759      Please bring American Airlines to #BlackBerry              0   \n",
       "13760  you have my money, you change my flight, and d...              0   \n",
       "13761  we have ppl so we need know how many seats are...              0   \n",
       "\n",
       "      review_coordinates review_timestamp    review_city  \\\n",
       "0                    NaN  24/2/2015 11:35            NaN   \n",
       "1                    NaN  24/2/2015 11:15            NaN   \n",
       "2                    NaN  24/2/2015 11:15      Lets Play   \n",
       "3                    NaN  24/2/2015 11:15            NaN   \n",
       "4                    NaN  24/2/2015 11:14            NaN   \n",
       "...                  ...              ...            ...   \n",
       "13757                NaN  22/2/2015 12:01            NaN   \n",
       "13758                NaN  22/2/2015 11:59          Texas   \n",
       "13759                NaN  22/2/2015 11:59  Nigeria,lagos   \n",
       "13760                NaN  22/2/2015 11:59     New Jersey   \n",
       "13761                NaN  22/2/2015 11:58     dallas, TX   \n",
       "\n",
       "                    user_timezone  \n",
       "0      Eastern Time (US & Canada)  \n",
       "1      Pacific Time (US & Canada)  \n",
       "2      Central Time (US & Canada)  \n",
       "3      Pacific Time (US & Canada)  \n",
       "4      Pacific Time (US & Canada)  \n",
       "...                           ...  \n",
       "13757                         NaN  \n",
       "13758                         NaN  \n",
       "13759                         NaN  \n",
       "13760  Eastern Time (US & Canada)  \n",
       "13761                         NaN  \n",
       "\n",
       "[13762 rows x 13 columns]"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "22e39d58",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Univariate Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "72f9ff37",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e72a47a0bd0d42779111512c8b90eeb9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "                                             |                                             | [  0%]   00:00 ->…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "analyze_report = sv.analyze(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "35e122ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report report.html was generated! NOTEBOOK/COLAB USERS: the web browser MAY not pop up, regardless, the report IS saved in your notebook/colab files.\n"
     ]
    }
   ],
   "source": [
    "analyze_report.show_html('report.html')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "e9ef3467",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f6f3903dba1a49cfa2aa49d06c7238db",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "                                             |                                             | [  0%]   00:00 ->…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "analyze_reportchallenge = sv.analyze(challenge)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "3d049637",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report challenge.html was generated! NOTEBOOK/COLAB USERS: the web browser MAY not pop up, regardless, the report IS saved in your notebook/colab files.\n"
     ]
    }
   ],
   "source": [
    "analyze_reportchallenge.show_html('challenge.html')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "98f45878",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1.376200e+04\n",
       "mean     5.692199e+17\n",
       "std      7.793554e+14\n",
       "min      5.675883e+17\n",
       "25%      5.685594e+17\n",
       "50%      5.694844e+17\n",
       "75%      5.698917e+17\n",
       "max      5.703106e+17\n",
       "Name: review_id, dtype: float64"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['review_id'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2883fb6f",
   "metadata": {},
   "source": [
    "We need to first identify dependent values, independent values , categorical values, and continuous values before we intitial our EDA analysis. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "ae3fd7b8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11230</th>\n",
       "      <td>570286841737318400</td>\n",
       "      <td>djjohnpayne</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>0.6304</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>you guys are killing me. http://t.co/iPGeIcSm</td>\n",
       "      <td>0</td>\n",
       "      <td>[0.0, 0.0]</td>\n",
       "      <td>24/2/2015 10:19</td>\n",
       "      <td>Las Vegas, NV</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11384</th>\n",
       "      <td>570286841737318400</td>\n",
       "      <td>djjohnpayne</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>you guys are killing me. http://t.co/iPGeIcSm</td>\n",
       "      <td>0</td>\n",
       "      <td>[0.0, 0.0]</td>\n",
       "      <td>24/2/2015 10:19</td>\n",
       "      <td>Las Vegas, NV</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id    user_name airline_sentiment  \\\n",
       "11230  570286841737318400  djjohnpayne          negative   \n",
       "11384  570286841737318400  djjohnpayne          negative   \n",
       "\n",
       "       sentiment_confidence   negative_reason  negative_reason_confidence  \\\n",
       "11230                   1.0  Cancelled Flight                      0.6304   \n",
       "11384                   1.0  Cancelled Flight                      1.0000   \n",
       "\n",
       "      airline_name                                    review_text  \\\n",
       "11230       Qantas  you guys are killing me. http://t.co/iPGeIcSm   \n",
       "11384       Qantas  you guys are killing me. http://t.co/iPGeIcSm   \n",
       "\n",
       "       thumbup_count review_coordinates review_timestamp    review_city  \\\n",
       "11230              0         [0.0, 0.0]  24/2/2015 10:19  Las Vegas, NV   \n",
       "11384              0         [0.0, 0.0]  24/2/2015 10:19  Las Vegas, NV   \n",
       "\n",
       "                    user_timezone  \n",
       "11230  Pacific Time (US & Canada)  \n",
       "11384  Pacific Time (US & Canada)  "
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train[\"review_id\"]==570286841737318400]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "afccdef8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11298</th>\n",
       "      <td>570267562623152128</td>\n",
       "      <td>alinaxkristin</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>you could train your flight attendants to have...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 9:02</td>\n",
       "      <td>Miami</td>\n",
       "      <td>Quito</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11453</th>\n",
       "      <td>570267562623152128</td>\n",
       "      <td>alinaxkristin</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "      <td>0.6688</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>you could train your flight attendants to have...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 9:02</td>\n",
       "      <td>Miami</td>\n",
       "      <td>Quito</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id      user_name airline_sentiment  \\\n",
       "11298  570267562623152128  alinaxkristin          negative   \n",
       "11453  570267562623152128  alinaxkristin          negative   \n",
       "\n",
       "       sentiment_confidence              negative_reason  \\\n",
       "11298                   1.0  Flight Attendant Complaints   \n",
       "11453                   1.0  Flight Attendant Complaints   \n",
       "\n",
       "       negative_reason_confidence airline_name  \\\n",
       "11298                      1.0000       Qantas   \n",
       "11453                      0.6688       Qantas   \n",
       "\n",
       "                                             review_text  thumbup_count  \\\n",
       "11298  you could train your flight attendants to have...              0   \n",
       "11453  you could train your flight attendants to have...              0   \n",
       "\n",
       "      review_coordinates review_timestamp review_city user_timezone  \n",
       "11298                NaN   24/2/2015 9:02      Miami          Quito  \n",
       "11453                NaN   24/2/2015 9:02      Miami          Quito  "
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train[\"review_id\"]==570267562623152128]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "02dff8ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "train=train.sort_values(by=['sentiment_confidence'])\n",
    "train=train.drop_duplicates(subset=['review_id'], keep='last')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db5a9b77",
   "metadata": {},
   "source": [
    "from the above example we can see several values are duplicated with the only difference being the negative_reason_confidence, which made the review_id not an unique key.\n",
    "Therefore, we will remove the values with lower sentiment confidence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "f74c9bbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = train.drop_duplicates(subset='review_id', keep=\"first\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d4340fe",
   "metadata": {},
   "source": [
    "Changed sentiment analysis from words into numeric values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "38de65cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>thumbup_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.362000e+04</td>\n",
       "      <td>13620.000000</td>\n",
       "      <td>9784.000000</td>\n",
       "      <td>13620.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.692091e+17</td>\n",
       "      <td>0.900895</td>\n",
       "      <td>0.639702</td>\n",
       "      <td>0.083480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.760599e+14</td>\n",
       "      <td>0.162454</td>\n",
       "      <td>0.330721</td>\n",
       "      <td>0.766913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.675883e+17</td>\n",
       "      <td>0.335000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.685521e+17</td>\n",
       "      <td>0.693200</td>\n",
       "      <td>0.360900</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.694595e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.670700</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.698832e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.703106e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>44.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          review_id  sentiment_confidence  negative_reason_confidence  \\\n",
       "count  1.362000e+04          13620.000000                 9784.000000   \n",
       "mean   5.692091e+17              0.900895                    0.639702   \n",
       "std    7.760599e+14              0.162454                    0.330721   \n",
       "min    5.675883e+17              0.335000                    0.000000   \n",
       "25%    5.685521e+17              0.693200                    0.360900   \n",
       "50%    5.694595e+17              1.000000                    0.670700   \n",
       "75%    5.698832e+17              1.000000                    1.000000   \n",
       "max    5.703106e+17              1.000000                    1.000000   \n",
       "\n",
       "       thumbup_count  \n",
       "count   13620.000000  \n",
       "mean        0.083480  \n",
       "std         0.766913  \n",
       "min         0.000000  \n",
       "25%         0.000000  \n",
       "50%         0.000000  \n",
       "75%         0.000000  \n",
       "max        44.000000  "
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "24cbaf15",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Histogram of ApplicantIncome'}, xlabel='thumbup_count'>"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAggAAAHBCAYAAAAM80OCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAApoUlEQVR4nO3df3zO9eL/8ed1bWPX2O+ZMC3k5yRE+ZSQUkKs/BijVsPBEKlzOqfzrcip06mck9Q5/VD6iVbKj1SS/Cil5BDJSZSlxCZjYz+a7fX9Q9c789pmZrNredxvt93adV2v63299n692/XY+7o2LmOMEQAAwHHc1T0BAADgewgEAABgIRAAAICFQAAAABYCAQAAWAgEAABgIRAAAICFQAAAABYCAQAAWAgEVLoePXrI5XJp1apV1m3PP/+8XC6Xbr75ZklSdna2WrRoodatW+vw4cNndqK/Izt27FDfvn0VHh6usLAw/fjjj2WO/+KLL+RyueTxeHTo0KEqndvNN98sl8ul119/XZL0+uuvKyoqSn/605+q9HFPxapVq+RyudSjR4/qngrgMwgEVKs6deooJSVF48aNU1BQULnvFx8fX2qEnI1GjBiht99+W23btlVCQoICAwPLHD9nzhxJUl5enubPn38mpujo2LGjUlJS1K9fvzP6uF4cO0D5+Ff3BHB2c7vdmjx5cnVP46SOHj0qf3/f/d/l888/V+3atbV69Wq53WV3f0FBgebOnauoqChlZmbqhRde0JgxY87QTKWmTZvqvvvuO2OPB6BiOIOAaudyuVS3bl3nckFBgaZOnapmzZopKChIrVq10owZM1RUVKRdu3bJ5XJp0aJFkqQrrrhC5513nnPfb775RoMGDVJkZKQ8Ho+6du2q5cuXF3u87OxsjRw5UuHh4QoODtawYcOUnJxc7KdK78sk06ZNU9OmTdW8eXNJUlFRkf75z38qLi5OgYGBio2N1eTJk5Wdne1s/7zzzlOtWrU0f/58tWnTRrVr11aLFi20ePFirV27Vpdccok8Ho+aNm2qJ554osx9U1BQoGnTpqlZs2aqXbu2mjRpoqlTp6qgoEDSby/ZFBYWKj8/X35+fic9Tf72228rIyNDI0aM0FVXXaVPPvlE27dvLzZm6tSpcrlcmj59uq6++mqFhISoUaNGmj59urz/vpt3La688krdcccdaty4serUqaMBAwbohx9+KPXxvXOeMGGCc93hw4c1efJkxcTEKCgoSO3atdOzzz5b7H6ffPKJevfurZCQEIWHh6tXr17asGGDNed//etfGjp0qEJCQhQdHa0JEybol19+OemxcyLvOr777rvq0KGDPB6PWrdurYULFxYbt2HDBvXu3VvBwcGKjIxUnz59tGXLlmJjli1bpssvv1x169ZVWFiYBg4cqB07dlj75LbbbtPw4cMVFham4OBgJSYm6ueff9aUKVMUHR2t0NBQ9enTR99//32x7b///vu67LLLFBQUpOjoaI0ZM6bYMQlUiAEqWffu3Y0ks3LlSuu2OXPmGEkmKSnJuU6SqVOnjnN5/PjxRpJp3bq1SU5ONo0bNzaSzN13321+/vlnM378eNOkSRMjycTHx5t77rnHGGNMWlqaCQ8PN5JMnz59zA033GD8/f2N2+02S5YscbZ/ww03GEkmNjbW3HjjjaZly5ZGUrE5e7+G2rVrm8GDB5s777zTGGPM7bffbiSZ5s2bm5EjR5oLL7zQSDKJiYnO9mNjY40kExYWZhISEkzPnj2NJOPxeEydOnXMddddZ4YOHWpq1aplJJmPP/641H05cOBAI8m0bNnS3HjjjSYmJsZIMkOGDDHGGLNu3TpnfwUEBJjx48ebRx99tMz1iY+PN5LM+vXrzYsvvmgkmb/85S/Fxtx7773OPunWrZsZOnSoqVu3rpFkHnvsMWOMMd99950zJjY21owYMcLZlxdddJGzraSkJCPJvPbaa8WOgfHjxxtjjCksLDTdunUzkkz79u1NUlKSqV+/frHH+vzzz02tWrWMx+MxiYmJ5vrrrzdut9tERESYn376qdic/fz8zBVXXGFuvvlm53h4/PHHyzx2Vq5caSSZ7t27W+sYGhpqRowYYfr06WMkmaCgIJOenm6MMWbz5s3G4/EYPz8/Ex8fbwYMGGBcLpcJCwszaWlpxhhjFi1aZNxut6lbt64ZOnSo6dq1q5FkoqKinDHefSLJdO7c2SQlJZmoqChnXGxsrElOTjYXXHCBkWSuvvpqZ55LliwxbrfbREZGmqSkJGdfXn/99WUeB8DJEAiodN4n17I+ygoEbxBkZGQYY4zZvn27iYuLM/369XPGDBgwwIqQW265xUgy06dPd657+eWXndgwxpgtW7YYSaZ+/frmwIEDxhhjcnNznSf6EwNh7ty5xb62KVOmmKSkJLN//35jjDE5OTkmODjYBAYGmqNHjxpjfntiOf6J/9JLLzWSTEpKinPdX//6V2u+x1u9erWRZNq2bWtycnKMMcbs37/fefL89NNPS92HpcnIyDABAQGmRYsWxhhjsrOzTVBQkImJiTGFhYXOOO+T7ejRo53rvE+ijRo1Msb8FgjH78ucnBzTunVrI8m89957xpiTB8LixYuNJNO1a1dnH3733XfGz8/PNG/e3BhjzBtvvGGSkpLMwoULnfmMGjXKSDIvv/xysTnfcsstzpjU1FQjyQwePNi5rqRjp6xAWL16tXOdNxKWLl1qjPktNp966ilnzPTp040kc//99xtjjGnWrJmRZD788ENr7uPGjSu2T6655hpTVFRkjDHmvffec4Jk165dznoFBgaagIAAZ72aNWtmAgICzPbt253t9+vXz0gyW7ZsMUBF+e6Lqqjx4uPj1ahRo2LX/e9//9OKFSvKvF/nzp21e/duTZ48WcnJybrsssv05ZdfnvTxli1bJkkaO3asc11iYqImTpyobdu2KTMzU+vXr5ckDRo0SOHh4ZKkwMBAnXfeefriiy+sbXpfWvCaMWOGfvrpJ6Wmpur77793TuPm5eXp4MGDioyMLPG+zZs318cff6wrrrjCue7888+XJO3bt6/MrycpKUkej0eSFBkZqaFDh2rmzJlau3atLr744pPul+PNnTtXBQUFGj58uCSpbt266t+/v+bPn68VK1aoV69exca3aNHC+bxHjx5q1qyZdu7cqYyMDOf6qKgoZ196PB4lJibq7rvv1qZNm6ztlWT16tWSpKFDh8rPz0/SsdP7K1eudF5Kuf7663Xddddp8eLFmj59ujIyMpz12rt3b7HttWnTxvncu4+Pn++pKmt73rknJiY6Y1JSUtSlSxfFxMTom2++0c6dOxUXF6euXbs6Y8aPH6/Zs2dr7dq1xR7r/PPPl8vlkvTb8dOkSRPFxsZKOrZe9evXV1pamjIzM5WZmamdO3cqIiJCM2fOdLbj3SdbtmxR27ZtK/y14+xGIKDKTJo0yXo9/Pnnnz9pIMyZM0cxMTF68cUX9corr8jj8WjQoEF66KGHdM4555R6v/T0dAUEBCgqKsq5zuVyqX79+srMzNThw4eVnp4uSWrYsGGFvqbZs2dr/Pjx+uWXX6zbzK+vzZ+q0u5X2lwbNGggSRX6tdDnn39ekrRixQpt3rxZkrRz507ntpM9oderV087d+7U3r17FRwcXOoYyX7iLk1mZqYkFVs3Sbr88sudz3ft2qUrr7xS3377rXX/sva798m2omtzsu1lZmYqMDCw2HtoIiIidNVVV0mSEwCVuYZexhhnHx84cKDE97P89NNPFd4+wJsU4XNCQkI0c+ZM7d+/X+vXr1dycrJeeuklDR48uMz7RUREqKCgwHnCkX77JupyuVSvXj3nG3VpP7WXJT8/X5MnT1ZRUZHmzZunrKwsGWOcn+4qW0REhCR7rnv27JEk1a9f/5S2t2XLFm3cuFGStGbNGi1YsEALFizQpk2bJElvvvmmsrKyytyG9wmnrFArz5jjeb/O/fv3F7s+JydHR44ckSTdf//9+vbbb3XjjTdq9+7dKiws1L333luu7VeliIgI5eXlOfOUpMLCQh0+fFj5+fmVvoYnCgkJkSS1bNlS5thLxsU+pkyZclrbx9mNQIBP2bZtm+rWravOnTvL5XKpU6dOmjFjhvz9/Yu9BOD9lUPvKWhJzk9tzzzzjHPd/PnzdfDgQXXt2lWBgYHq1KmT80d7vH8g6OjRo9a7wkuSnZ2tI0eOKDo6WgkJCQoODtbBgwdP+qRaUd6v56WXXlJ+fr6kYz+xpqamOr89cCq8Zw/+8Y9/WE8kN998s3Jzc5WamlrsPrt27XI+/+CDD5SWlqaYmBjnLIF07Ind+5Nwbm6u5s2bJ+nY3zsoj+7du0uSUlNTVVhYKOnY2ZOIiAjnlL43OoYPH66YmBi53e5yrVlJSjp2Kso7d+/XLEmPP/64goODNXXqVLVq1UoxMTHavHmzPv30U2fMU089Jem3Na6ouLg4RUdH6+uvv9bixYud6/ft26fZs2ef1rYBAgE+pXXr1urcubM+//xzXXLJJfrDH/6gLl266OjRo+rbt68zzvv67K233qqUlBRJx37NLSQkRH/+85/Vr18/DRo0SElJSapVq5YeeughZ/tDhgzRnj171L59eyUnJ6tDhw7OT9ZliYqKUseOHbVnzx51795dw4cPV6tWrZwzFnl5eZW6L3r16qU+ffpo48aN6tChg5KSktS+fXvt3btXEyZMULNmzcq9raNHj+qVV16RJN1www3W7QMHDpT0W0R4PfHEE+rdu7eGDx+u/v37S5LuuuuuYmP27dun9u3ba+TIkerQoYO2b9+uiy++uNwB06dPH3Xr1k1r1qxR586dlZycrI4dOyo/P1933HGHJOmaa66RJCUnJys5OVmdOnUq9seeTkVJx05F3XPPPfJ4PBo3bpwGDhyoQYMG6Y477lBoaKhGjx4tl8ulRx55RC6XS7169VJiYqJ69Oihp556Sueee+5p/4Tv5+enGTNmSDq2rvHx8brpppvUpk0bTZo0yTlTAVQEgQCfs2jRIt12223au3evXnzxRR06dEh//OMfi/1ENHnyZPXo0UNpaWnOa+neNwL27dtXa9as0bJly9StWzetWrVKXbp0ce47e/ZsjRo1SpmZmVq6dKm6dOnivN4dEBBQ5tzeeOMN9e/fX5s3b9ZHH32klJQU53X7bdu2Vfau0IIFC3TnnXcqOztb8+fPV2BgoB555JFib0grj3fffVf79u1Tu3btnJ/Kj9erVy8FBwdr7dq1xX4/PyUlRW63W0uWLFFISIj+9re/ady4ccXu27JlSw0dOlTLly/X7t271b9/f7355pvlnpvb7dbSpUs1adIk7du3T6+88orCwsL09NNP6/bbb5ckTZgwQdOmTZOfn58WLFighg0bOtF3qvu9pGOnotq1a6cPP/xQPXv21LJly7Rs2TJdffXV+vDDD9W0aVNJUkJCghYuXKhWrVrpjTfe0NatWzVixAh98sknCgsLO63Hl479Fc233npLl1xyid577z0tWrRIF198sT788MMKv9cGkCSXqax37wA1xI4dO+Tv7+/8kZysrCy1bNlS6enp2rt3b7HT52erqVOnatq0aXr44Yedn+JPtGvXLjVp0kRxcXHl+i0TADULv8WAs05iYqK++uor9ezZU5GRkXr//fe1d+9epaSkEAcA8CteYsBZ5/XXX9eQIUP02Wef6bXXXlNkZKT+/e9/a9asWdU9NQDwGbzEAAAALJxBAAAAFgIBAABYCAQAAGCp8G8xFBUVKTPzgAIDPc7fJwcAAL7NGKO8vFyFh0fI7S79PEGFAyEz84AmjB1V0bsDAIBq9PiTsxUZGVXq7RUOhMBAj/MAHk9QRTcDAADOoNzcHE0YO8p5Hi9NhQPB+7KCxxOkoCACAQCAmuRkbw/gTYoAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACwEAgAAMBCIAAAAAuBAAAALAQCAACw+Ff3BI5njFF+fr4kqXbt2nK5XNU8IwAAzk4+dQYhPz9f8fHxio+Pd0IBAACceT4VCHl5eSV+DgAAziyfCgQAAOAbCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgMWnAqGoqKjEzwEAwJnlU4GQnZ1d4ucAAODM8qlAAAAAvoFAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgKVGB0Lv3r2dD18cJ0m33367evfurdtvv73McQ888IB69+6tBx54oFLGrVu3TjfddJPWrVt30jmeytjyqOztVYXqnGNN2D8vvPCC+vTpoxdeeKG6p3LG1IRjoiYcO6gcvrDWNTYQTnxyLu3Jul+/fmVe9po1a1aZl73mzZtX5uXj7d69W1u3bpUkbd26Vbt37y5xXHp6utasWSNJWrNmjdLT009rXF5enmbNmqX09HTNmjVLeXl5pc7xVMaWR2VvrypU5xxrwv45dOiQ5s+fr6KiIs2fP1+HDh2q7ilVuZpwTNSEYweVw1fWusYGQnkdPXq0zMteS5cuLfOy14k/UZX1E9att95a7PKkSZNKHDdlypRil0s721Deca+++qoOHDggSTpw4IBSU1NLneOpjC2Pyt5eVajOOdaE/XPffffJGCNJMsZo+vTp1TyjqlcTjomacOygcvjKWtfIQCjtbEF5zyqceP11111X4rgTr09ISChxXEnXp6amKjc3t9h1OTk51kIvX75c+/fvL3ZdRkaGli9fXqFxe/bsUWpqarFv8KmpqdqzZ481x1MZWx6Vvb2qUJ1zrAn7Z+PGjc5ZL68vv/xSGzdurKYZVb2acEzUhGMHlcOX1rrcgVBQUKCcnBznIzc3pyrnVaryvD9AknMqvjTe2/fv36+CgoISxxQUFDhPyocOHSr1VOuJtx09elTPPfdciWOfe+455yxGYWGhHn300RLHPfrooyosLDylccYYPfHEE9YY7/XeA+5Ux5ZHZW+vKlTnHGvC/ikqKtLf//73Em/7+9//rqKiojM8o6pXE46JmnDsoHL42lqXOxAWvfm6RiYlOh8Txo6qynmdtvK82U+SkpKSyhznvf2WW24pc9zxt8+dO7fMsd7b33nnHefJ/USFhYV65513Tmnc7t27tWHDBmtsYWGhNmzYUOw9EKcytjwqe3tVoTrnWBP2z/r165WVlVXibVlZWVq/fv0ZnlHVqwnHRE04dlA5fG2tyx0IA64fpGdfmOt8PP7k7Kqc12m76667ynX7yd6l7b19zpw5ZY47/vbExMQyx3pvv/baa+Xn51fiGD8/P1177bWnNK5x48a66KKL5Ha7rTGdOnVS48aNnetOZWx5VPb2qkJ1zrEm7J/OnTsrJCSkxNtCQkLUuXPnMzyjqlcTjomacOygcvjaWpc7EAICAhQUFOR8eDxBVTmvUr377rvlur1bt25ljvPeHhUVpYCAgBLH1KpVS1FRUZKk0NBQhYaGljguPDy82G3+/v5KTk4uceyoUaPk7+8v6diiT548ucRxU6ZMcaKgvONcLpfGjx8vl8tVbIzL5VJKSkqx609lbHlU9vaqQnXOsSbsH7fbrb/85S8l3nbXXXdZ37R+D2rCMVETjh1UDl9b6xr5f3xpkXDi9eUdt2TJkhLHLV68uNjlV199tcRxJf2q45AhQ+TxeIpdFxQUpEGDBhW7rlevXk6EeNWrV09XXnllhcY1bNhQQ4YMKfbNZciQIWrYsKE1x1MZWx6Vvb2qUJ1zrAn7p0OHDoqLiyt2Xdu2bdW+ffvqmdAZUBOOiZpw7KBy+NJa18hAOBXen9ZLu+zVt2/fMi97nfiehbLew/DYY48Vuzxz5swSx/3zn/8sdnnGjBmnNS4hIUERERGSpMjISA0ZMqTUOZ7K2PKo7O1VheqcY03YP/fcc4/zzcntduvuu++u5hlVvZpwTNSEYweVw1fWusYGQnnPFrz11ltlXvaaOHFimZe9hg0bVubl4zVu3Nj5aSwuLq7U14+io6Odlzy6deum6Ojo0xoXGBioiRMnKjo6WhMmTFBgYGCpczyVseVR2durCtU5x5qwf0JDQzV06FC53W4lJCSU+tLa70lNOCZqwrGDyuEra+0yFfy9iZycHI1MStSzL8xVUFDlvB8hLS1NY8aMkSQ99dRTio2NrZTtAgCAY8r7/F1jzyAAAICqQyAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsPhUIAQHB5f4OQAAOLN8KhDcbneJnwMAgDOLZ2EAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFgIBAABYCAQAAGAhEAAAgIVAAAAAFp8KhMDAwBI/BwAAZ5Z/dU/geLVr19bChQudzwEAQPXwqUBwuVycOQAAwAf41EsMAADANxAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsBAIAADAQiAAAAALgQAAACwEAgAAsPhX9I7GGElSbm5OpU0GAABULe/ztvd5vDQVDoS8vFxJ0oSxoyq6CQAAUE3y8nJVp06dUm93mZMlRCmKioqUmXlAgYEeuVyuCk/wRLm5OZowdpQef3K2PJ6gStsuTh9r45tYF9/F2vims31djDHKy8tVeHiE3O7S32lQ4TMIbrdbkZFRFb37SXk8QQoKOvsWriZgbXwT6+K7WBvfdDavS1lnDrx4kyIAALAQCAAAwOJzgRAQEKCBgxMUEBBQ3VPBCVgb38S6+C7WxjexLuVT4TcpAgCA3y+fO4MAAACqH4EAAAAsBAIAALBU+O8gVJVP1n6k+XNf0pEjh9Xp4i4aOXosbySpJocOHdTaj9ZozaqVGj9xshqfG6v9GRn6zxMz9e3OHWrQsJHGpExUbOx51T3Vs8qSRW/q7bcWKz8/T+3ad9DYlFt1ODubdalmhYWFenXey/rg/fckSR06dtLoseOVdegQa+MjHp3xkD5d97HmvbaQ72Xl4FNnELKzs/Tkvx/T9YOGaOr0B/XFpv9q5Yrl1T2ts1Jubq4mjvuD1q1dq7Rd3znXv/zSHNWuHaiH/zlL554bq6f/83g1zvLss/mLTXp13itKmThZ0+7/h77etk3vvr2EdfEBH324Wh+8v1x//PP/07T7/6Evt2zWB++/x9r4iM/Xf6b1n61zLrMuJ+dTgbBzxzcyRureo6diGjdWhw4XadtXW6t7WmelWrVqadZ/ntbEyVOKXb/tq63q2q27ourVU4+eV+nbnTuUn59fTbM8+/j7+ytxxE26oN2Fatz4XDVo2FCHDh5iXXxA9x49Nfv5l9WyVWuFhYXJ399ffn5+rI0PyMnJ0Zxnn9I11/Z1rmNdTs6nAiErK0u1A2s7/7ZDUJ06ysrKquZZnZ38/PwUGhpmXZ+dlaXAwEBJv/2pzmzW6IxpE9dWffr1lySl7fpOO77Zrssu78a6+JA/3T5Jf0i+Sec0aKCrru7N2viAeS+/qAvatVfbC9o517EuJ+dTgVCSSvx3oFBFWKMz7+ef9+vhB+9X334DdH7zFiWOYV2qx5133a2//L97lbZrl1av+qDEMazNmfO/bV/p8/WfasSNN590LOtSnE+9STE4OFh5ubkqKiqS2+1Wbk6OQkr4KRbVJzg4REeOHJF07H0KkhQcElqdUzrrZGVl6YHpUxXX9gINGTZcEuviC9LSdunQwYNqd2F7RUZGqU3btvpq65esTTV7c8Frys7O1qQJY1RYWCRJGpmUyLqUg0+dQWh2fgu53W6tWL5MP/74gzZt+q/axLWt7mnhOK3bxOnD1Su1PyNDq1auUPPmLVWrVq3qntZZIy83Vw89MF1RUVEakXSLcnKOKCcnh3XxAd/v+k7/euRBfbP9a+3e/b2+2f61zmvShLWpZuPG36p/PfaEHnz4X0pKHiVJevDhf7Eu5eBzf2r5k7Ufad4rLyon54g6X9xFyfyaY7XKSN+nW8eP0UMzZh73a46PaeeOb9SwUSONG3+rGp8bW93TPGusXvWBnnzisWLXRdWrp3unPcC6VDNjjF556XmtWbVSxhh1ufQyJd0ySgczM1kbH/HfDev18IP3H/drjqxLWXwuEAAAQPXzqZcYAACAbyAQAACAhUAAAAAWAgEAAFgIBAAAYCEQAACAhUAAzoDXU+dp2OB4ZaTvOyOPNzFltO67969n5LEA/D4RCEAV+c/jMzVscHx1T+OsMWxwvP7z+Mzqngbwu0EgAAAAC39JEagCE1NGa39GhnO5dZs4bftqqxJHJOmD999TdnaWunbroZuTRzvj69WL1j3T7pckrV65Qk/+e5bunjpdbeIu0LDB8br0ssuVl5enbV99qdDQMI1JmaD1n36qVSvfV0BALfW9boCuG3C9s726dYNV/5xztHnTRoVHRGr0mHFq1TpOX23doulT79bYlInqfsWVkqT77v2rMjLSNevfzziPPeD6gVr70Rodzs7WZZd31y0j/yA/P78yv+70ffs059mn9dXWLapTp64u79ZDg4cmyt/fX0ePHlXq/Fe0ZtVK5ebmqE1cW90ycoyi69d3/qT3wMEJGjRkmKRjZ2DWrF6pea8tdOY8cHCCNny+Xnt+/EEtWrbWxElT9MMP32v61LudOUTVq6dZ/36m8hYTOEtxBgGoAgnDRqh585aSpOTRY9WgQUNJ0qqVK9Sj51VqfO55WvbOUm3Z/EW5t/nJx2sVHV1f1/btr59/3q+/TbtXO3Zs1w2DEhQaFqZ5r7yon/b86Izf9d23kqT+8TcoK+uQHp3xsAoKCsr/eGs/0pVXXa1WrdtoxfJlWr7snTLHFxUV6ZGHHtCOHdsVf8NgxbW9QIsXvaHFCxdIkl564TktWfSmLurUWf2ui9e2r7bqoQf/pqNHj5Z7Tm+/tUQdO3bSZZd315dbvtDSJYt0ToNGSh49VpLUvHlLJQwbUe7tASidT/1zz8DvRdfLu2vLF5v0zTdfq9fVvfV66jxJ0q233aHY2PPUvuNF+vMdk5W26ztd0O7Ccm2zy/9d6vxrdD/sTtNnn67T5Nv/pPDwCJ3ToIEe+ccDSkvbpQYNG0mSYs9roslT/iRJioiM0n8en6lNGzeoTp065Xq8W0aNUfsOHVVUWKhJE8Zq1coV6t2nX6njN3+xSbu/T1PKxMm6vFsPSVJ6+j5t//pr5ebmauWK5bqky6UaPXa8JCk0LExznn1amzb+V7Gx5ftHcgYnDNO1fa+TJK3/bJ2+/z5NERER6nV1bz33zJNq0LChul7evVzbAlA2AgE4g4I8nmL/zcvLLfd9j/9XTQMDj90/PDxCkuT5dXv5eXm/PVZQkPN506bnS5Iy0tNVp0mTcj1erVrHHs/t56dm5zfX5i82lTl+z48//PpYzZzrpv3tQUnSzh3fqKCgQK1at3Fu836+Z88P5Q6E47+mwMBA5efnlTEawOngJQbgLOByHftvRd9yVFRUJOkk93WVfpP/r3Hjdv/2Lcc7F7eLb0OAL+L/TKCKuH59Vj725Fo2T6BHOUeOOJdzcnJO+/Hzcn87O+F9P0J0/frO2YcjOb89Xm6ufSbjl19+kXRs/t/u3KHo+ueU+XgNG8YUeyxJevaZJzVn9tM655wGCggI0P/+t8257etfP298bqwCfz0DcuQ09wHvuQYqDy8xAFUkNDRM0rE352UdOlTm2KbnN9fqlSv0/HPPyM/tpw9WvHfaj//dd9/qP4/PVMNGMXr7rcUKCwtX+w4XqfDoUXk8QVq6ZJHy8vK0Y/t27fruW0XVq1fs/s8985R6XtVL27/+Wj//vF/9+seX+Xjt2l2omMbn6oU5zyojI117f/pJq1d9oPgbBqt27dq6pndfvbVkoYI8HoWFh2vpkkVqfG6s2l7QTn5+foqOrq81qz5Qnbp1tfenn/T5+k9P6esNDQ3T1i+3aOmSRep73YBT3V0ATsAZBKCK9Ox1tWIan6vVK1coM/NAmWOHJCQqru0FWvXBCm39crN6XXPtaT9+p84XK/+XfC1683XVDQ7WpCl/VEBAgAI9Ho0dP1F+bj8tXbxQgYGBantBO+v+HTp20soV7+urrV/qyl7XqNc11yo/P18//7zf+jiYmSm3n5/uuPMuNTv/fL254DVt/O8G9R9wgwYNGSpJShg2XL2v7adP132ipUsWqe0F7fTnu+5xfnVy3IRbFR4RqbeXLNLh7Gz936VdT+nrHTw0Ufn5eVr5wfunve8A8HcQAJzgxL/BUNJtJ+JvDwC/P7zEAKDc2ra7UH/8s/1vPNSqVbsaZgOgKhEIAMotMjJKkZFR1T0NAGcALzEAAAALb1IEAAAWAgEAAFgIBAAAYCEQAACAhUAAAAAWAgEAAFgIBAAAYCEQAACAhUAAAACW/w99VCG2UdBIfgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.set(title =\"Histogram \")\n",
    "sns.distplot(a=train[[\"thumbup_count\"]], bins=20)\n",
    "fig, ax = plt.subplots()\n",
    "ax.set(title =\"Histogram of ApplicantIncome\")\n",
    "sns.boxplot(x=\"thumbup_count\", data=train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "335397e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Histogram '}, ylabel='Density'>"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.set(title =\"Histogram \")\n",
    "sns.distplot(a=train[[\"sentiment_confidence\"]], bins=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "id": "ea1ec817",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>930</th>\n",
       "      <td>569972097453137920</td>\n",
       "      <td>cristobalwong</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.3350</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>Thank you, ^JH, appreciate the prompt response...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:28</td>\n",
       "      <td>San Francisco Bay Area</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5873</th>\n",
       "      <td>568092537786748928</td>\n",
       "      <td>VegasRenegade</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.3353</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>All Flights lead to VEGAS BABY!</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 8:59</td>\n",
       "      <td>The Wild Blue Yonder</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5963</th>\n",
       "      <td>568028183267639297</td>\n",
       "      <td>CourtSnod</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3356</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>de-icing is important!</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 4:44</td>\n",
       "      <td>Tennessee Original</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7436</th>\n",
       "      <td>568993773277069312</td>\n",
       "      <td>KoolieAshSays</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.3363</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>oh. Makes sense. My bad.</td>\n",
       "      <td>0</td>\n",
       "      <td>[40.88529232, -73.856635]</td>\n",
       "      <td>20/2/2015 20:41</td>\n",
       "      <td>ÜT: 17.9889591,-76.7712636</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2254</th>\n",
       "      <td>569227372223811584</td>\n",
       "      <td>zachkpearson</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>. Thanks. Hopefully this is easily resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 12:09</td>\n",
       "      <td>Portland, Oregon</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5281</th>\n",
       "      <td>568853498726449152</td>\n",
       "      <td>djknuckles</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>hello I was talking with but it says you moved...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 11:23</td>\n",
       "      <td>WORLDWIDE</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5282</th>\n",
       "      <td>568853440794730496</td>\n",
       "      <td>woman4lif</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>I am following you now.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 11:23</td>\n",
       "      <td>Northern UT</td>\n",
       "      <td>Mountain Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5283</th>\n",
       "      <td>568852204750135296</td>\n",
       "      <td>woman4lif</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>I can go. I want to badly!!!! I don't know how...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 11:18</td>\n",
       "      <td>Northern UT</td>\n",
       "      <td>Mountain Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5274</th>\n",
       "      <td>568858194447085568</td>\n",
       "      <td>LRC110979</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6564</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>yes but not with out an attitude and nasty com...</td>\n",
       "      <td>0</td>\n",
       "      <td>[41.93001697, -72.68496357]</td>\n",
       "      <td>20/2/2015 11:42</td>\n",
       "      <td>Rochester Hills, MI</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6880</th>\n",
       "      <td>569671231001092096</td>\n",
       "      <td>malbeckford</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6699</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>\"Considering?\" #Accessibility is an absolute M...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 17:33</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>13620 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               review_id      user_name airline_sentiment  \\\n",
       "930   569972097453137920  cristobalwong           neutral   \n",
       "5873  568092537786748928  VegasRenegade           neutral   \n",
       "5963  568028183267639297      CourtSnod          positive   \n",
       "7436  568993773277069312  KoolieAshSays           neutral   \n",
       "2254  569227372223811584   zachkpearson          positive   \n",
       "...                  ...            ...               ...   \n",
       "5281  568853498726449152     djknuckles           neutral   \n",
       "5282  568853440794730496      woman4lif           neutral   \n",
       "5283  568852204750135296      woman4lif           neutral   \n",
       "5274  568858194447085568      LRC110979          negative   \n",
       "6880  569671231001092096    malbeckford          negative   \n",
       "\n",
       "      sentiment_confidence         negative_reason  \\\n",
       "930                 0.3350                     NaN   \n",
       "5873                0.3353                     NaN   \n",
       "5963                0.3356                     NaN   \n",
       "7436                0.3363                     NaN   \n",
       "2254                0.3366                     NaN   \n",
       "...                    ...                     ...   \n",
       "5281                1.0000                     NaN   \n",
       "5282                1.0000                     NaN   \n",
       "5283                1.0000                     NaN   \n",
       "5274                1.0000  Customer Service Issue   \n",
       "6880                1.0000  Customer Service Issue   \n",
       "\n",
       "      negative_reason_confidence        airline_name  \\\n",
       "930                       0.0000            Emirates   \n",
       "5873                      0.0000           Lufthansa   \n",
       "5963                      0.0000           Lufthansa   \n",
       "7436                      0.0000  Singapore Airlines   \n",
       "2254                      0.0000            Emirates   \n",
       "...                          ...                 ...   \n",
       "5281                         NaN           Lufthansa   \n",
       "5282                         NaN           Lufthansa   \n",
       "5283                         NaN           Lufthansa   \n",
       "5274                      0.6564           Lufthansa   \n",
       "6880                      0.6699  Singapore Airlines   \n",
       "\n",
       "                                            review_text  thumbup_count  \\\n",
       "930   Thank you, ^JH, appreciate the prompt response...              0   \n",
       "5873                    All Flights lead to VEGAS BABY!              1   \n",
       "5963                             de-icing is important!              0   \n",
       "7436                           oh. Makes sense. My bad.              0   \n",
       "2254       . Thanks. Hopefully this is easily resolved.              0   \n",
       "...                                                 ...            ...   \n",
       "5281  hello I was talking with but it says you moved...              0   \n",
       "5282                            I am following you now.              0   \n",
       "5283  I can go. I want to badly!!!! I don't know how...              0   \n",
       "5274  yes but not with out an attitude and nasty com...              0   \n",
       "6880  \"Considering?\" #Accessibility is an absolute M...              1   \n",
       "\n",
       "               review_coordinates review_timestamp  \\\n",
       "930                           NaN  23/2/2015 13:28   \n",
       "5873                          NaN   18/2/2015 8:59   \n",
       "5963                          NaN   18/2/2015 4:44   \n",
       "7436    [40.88529232, -73.856635]  20/2/2015 20:41   \n",
       "2254                          NaN  21/2/2015 12:09   \n",
       "...                           ...              ...   \n",
       "5281                          NaN  20/2/2015 11:23   \n",
       "5282                          NaN  20/2/2015 11:23   \n",
       "5283                          NaN  20/2/2015 11:18   \n",
       "5274  [41.93001697, -72.68496357]  20/2/2015 11:42   \n",
       "6880                          NaN  22/2/2015 17:33   \n",
       "\n",
       "                     review_city                user_timezone  \n",
       "930       San Francisco Bay Area                          NaN  \n",
       "5873        The Wild Blue Yonder       Atlantic Time (Canada)  \n",
       "5963          Tennessee Original   Central Time (US & Canada)  \n",
       "7436  ÜT: 17.9889591,-76.7712636   Pacific Time (US & Canada)  \n",
       "2254            Portland, Oregon   Pacific Time (US & Canada)  \n",
       "...                          ...                          ...  \n",
       "5281                   WORLDWIDE   Pacific Time (US & Canada)  \n",
       "5282                 Northern UT  Mountain Time (US & Canada)  \n",
       "5283                 Northern UT  Mountain Time (US & Canada)  \n",
       "5274         Rochester Hills, MI   Eastern Time (US & Canada)  \n",
       "6880                         NaN                          NaN  \n",
       "\n",
       "[13620 rows x 13 columns]"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "8073ceb6",
   "metadata": {},
   "outputs": [],
   "source": [
    "train['review'] = train.apply(lambda row: row['negative_reason'] if not pd.isna(row['negative_reason']) and row['negative_reason'] != '' else row['airline_sentiment'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "d9234de9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>review</th>\n",
       "      <th>Bad Flight</th>\n",
       "      <th>Can't Tell</th>\n",
       "      <th>Cancelled Flight</th>\n",
       "      <th>Customer Service Issue</th>\n",
       "      <th>Damaged Luggage</th>\n",
       "      <th>Flight Attendant Complaints</th>\n",
       "      <th>Flight Booking Problems</th>\n",
       "      <th>Late Flight</th>\n",
       "      <th>Lost Luggage</th>\n",
       "      <th>longlines</th>\n",
       "      <th>neutral</th>\n",
       "      <th>positive</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>airline_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>EgyptAir</th>\n",
       "      <td>18</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>57</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>16</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>155</td>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Emirates</th>\n",
       "      <td>208</td>\n",
       "      <td>351</td>\n",
       "      <td>166</td>\n",
       "      <td>637</td>\n",
       "      <td>20</td>\n",
       "      <td>156</td>\n",
       "      <td>132</td>\n",
       "      <td>492</td>\n",
       "      <td>261</td>\n",
       "      <td>46</td>\n",
       "      <td>660</td>\n",
       "      <td>465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lufthansa</th>\n",
       "      <td>82</td>\n",
       "      <td>152</td>\n",
       "      <td>149</td>\n",
       "      <td>367</td>\n",
       "      <td>13</td>\n",
       "      <td>37</td>\n",
       "      <td>60</td>\n",
       "      <td>146</td>\n",
       "      <td>87</td>\n",
       "      <td>27</td>\n",
       "      <td>623</td>\n",
       "      <td>537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Qantas</th>\n",
       "      <td>78</td>\n",
       "      <td>176</td>\n",
       "      <td>218</td>\n",
       "      <td>693</td>\n",
       "      <td>10</td>\n",
       "      <td>77</td>\n",
       "      <td>117</td>\n",
       "      <td>223</td>\n",
       "      <td>137</td>\n",
       "      <td>31</td>\n",
       "      <td>404</td>\n",
       "      <td>293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Qatar Airways</th>\n",
       "      <td>97</td>\n",
       "      <td>228</td>\n",
       "      <td>175</td>\n",
       "      <td>773</td>\n",
       "      <td>10</td>\n",
       "      <td>118</td>\n",
       "      <td>116</td>\n",
       "      <td>425</td>\n",
       "      <td>143</td>\n",
       "      <td>47</td>\n",
       "      <td>361</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Singapore Airlines</th>\n",
       "      <td>59</td>\n",
       "      <td>172</td>\n",
       "      <td>49</td>\n",
       "      <td>181</td>\n",
       "      <td>8</td>\n",
       "      <td>58</td>\n",
       "      <td>41</td>\n",
       "      <td>255</td>\n",
       "      <td>52</td>\n",
       "      <td>14</td>\n",
       "      <td>677</td>\n",
       "      <td>508</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "review              Bad Flight  Can't Tell  Cancelled Flight  \\\n",
       "airline_name                                                   \n",
       "EgyptAir                    18          21                17   \n",
       "Emirates                   208         351               166   \n",
       "Lufthansa                   82         152               149   \n",
       "Qantas                      78         176               218   \n",
       "Qatar Airways               97         228               175   \n",
       "Singapore Airlines          59         172                49   \n",
       "\n",
       "review              Customer Service Issue  Damaged Luggage  \\\n",
       "airline_name                                                  \n",
       "EgyptAir                                57                4   \n",
       "Emirates                               637               20   \n",
       "Lufthansa                              367               13   \n",
       "Qantas                                 693               10   \n",
       "Qatar Airways                          773               10   \n",
       "Singapore Airlines                     181                8   \n",
       "\n",
       "review              Flight Attendant Complaints  Flight Booking Problems  \\\n",
       "airline_name                                                               \n",
       "EgyptAir                                      5                       27   \n",
       "Emirates                                    156                      132   \n",
       "Lufthansa                                    37                       60   \n",
       "Qantas                                       77                      117   \n",
       "Qatar Airways                               118                      116   \n",
       "Singapore Airlines                           58                       41   \n",
       "\n",
       "review              Late Flight  Lost Luggage  longlines  neutral  positive  \n",
       "airline_name                                                                 \n",
       "EgyptAir                     16             3          1      155       142  \n",
       "Emirates                    492           261         46      660       465  \n",
       "Lufthansa                   146            87         27      623       537  \n",
       "Qantas                      223           137         31      404       293  \n",
       "Qatar Airways               425           143         47      361       256  \n",
       "Singapore Airlines          255            52         14      677       508  "
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(train['airline_name'], train['review'])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1ea407d",
   "metadata": {},
   "source": [
    "# Bivariate Analysis \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "93dd966f",
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_data_number = train[train['user_name'].str.contains(r'\\d', regex=True, na=False)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "29c64830",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "      <th>review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8810</th>\n",
       "      <td>569959415433965569</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6702</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "      <td>0.6702</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>I get some bs from a guy at the door says door...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 12:38</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9127</th>\n",
       "      <td>569694350831161344</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6718</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6718</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>always nice when a customer service manager th...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 19:04</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8771</th>\n",
       "      <td>569971542538346496</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6902</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.6902</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>the exhaustion of traveling n runnin thru the ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:26</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Can't Tell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8777</th>\n",
       "      <td>569970626057801728</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.7010</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>doesn't take into account the $ it cost rent a...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:22</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9138</th>\n",
       "      <td>569686602940858368</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.6596</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>why the hell did you overbook a plane when you...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 18:34</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bad Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9150</th>\n",
       "      <td>569682545375354880</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6632</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>we don't control your stupidity in not having ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 18:17</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Late Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9199</th>\n",
       "      <td>569660306298945537</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>this is the reason I avoid and like the plague...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:49</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Can't Tell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9198</th>\n",
       "      <td>569661113593425920</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.3481</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>how can a plane scheduled to come in on time a...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:52</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bad Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9202</th>\n",
       "      <td>569659137581326336</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>hire smarter IT people if your systems keep cr...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:44</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9204</th>\n",
       "      <td>569658553381883905</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6775</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>what is wrong with you guys and your inability...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:42</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Late Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9205</th>\n",
       "      <td>569658168097292288</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.6315</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>you guys have to be the worst. We are flying i...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:41</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Can't Tell</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8795</th>\n",
       "      <td>569967469135142914</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.3681</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>you idiots kept an on time plane on the Tarmac...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:10</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Late Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8794</th>\n",
       "      <td>569967808446009345</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6805</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>if you actually cared about what you did it wo...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:11</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8791</th>\n",
       "      <td>569968609440563201</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>is okay for u Cancelled Flight change a flight...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:14</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8805</th>\n",
       "      <td>569960904852312064</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>just an FYI I'll be writing registered letters...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 12:44</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8809</th>\n",
       "      <td>569959975566467072</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>I find it funny that responds but you don't. Y...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 12:40</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8745</th>\n",
       "      <td>569985288992264192</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.3561</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>PLANS CHANGED! IS THAT WHAT U CALL NOT BEIN AB...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 14:20</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8783</th>\n",
       "      <td>569969714245431296</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "      <td>0.6736</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>we had a friend there telling the gate attenda...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:19</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Flight Attendant Complaints</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8769</th>\n",
       "      <td>569972250385862656</td>\n",
       "      <td>ElmiraBudMan</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Lost Luggage</td>\n",
       "      <td>0.6723</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>only happened because u couldn't get us home t...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:29</td>\n",
       "      <td>Does it really matter</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lost Luggage</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               review_id     user_name airline_sentiment  \\\n",
       "8810  569959415433965569  ElmiraBudMan          negative   \n",
       "9127  569694350831161344  ElmiraBudMan          negative   \n",
       "8771  569971542538346496  ElmiraBudMan          negative   \n",
       "8777  569970626057801728  ElmiraBudMan           neutral   \n",
       "9138  569686602940858368  ElmiraBudMan          negative   \n",
       "9150  569682545375354880  ElmiraBudMan          negative   \n",
       "9199  569660306298945537  ElmiraBudMan          negative   \n",
       "9198  569661113593425920  ElmiraBudMan          negative   \n",
       "9202  569659137581326336  ElmiraBudMan          negative   \n",
       "9204  569658553381883905  ElmiraBudMan          negative   \n",
       "9205  569658168097292288  ElmiraBudMan          negative   \n",
       "8795  569967469135142914  ElmiraBudMan          negative   \n",
       "8794  569967808446009345  ElmiraBudMan          negative   \n",
       "8791  569968609440563201  ElmiraBudMan          negative   \n",
       "8805  569960904852312064  ElmiraBudMan          negative   \n",
       "8809  569959975566467072  ElmiraBudMan          negative   \n",
       "8745  569985288992264192  ElmiraBudMan          negative   \n",
       "8783  569969714245431296  ElmiraBudMan          negative   \n",
       "8769  569972250385862656  ElmiraBudMan          negative   \n",
       "\n",
       "      sentiment_confidence              negative_reason  \\\n",
       "8810                0.6702  Flight Attendant Complaints   \n",
       "9127                0.6718       Customer Service Issue   \n",
       "8771                0.6902                   Can't Tell   \n",
       "8777                0.7010                          NaN   \n",
       "9138                1.0000                   Bad Flight   \n",
       "9150                1.0000                  Late Flight   \n",
       "9199                1.0000                   Can't Tell   \n",
       "9198                1.0000                   Bad Flight   \n",
       "9202                1.0000  Flight Attendant Complaints   \n",
       "9204                1.0000                  Late Flight   \n",
       "9205                1.0000                   Can't Tell   \n",
       "8795                1.0000                  Late Flight   \n",
       "8794                1.0000       Customer Service Issue   \n",
       "8791                1.0000             Cancelled Flight   \n",
       "8805                1.0000       Customer Service Issue   \n",
       "8809                1.0000       Customer Service Issue   \n",
       "8745                1.0000       Customer Service Issue   \n",
       "8783                1.0000  Flight Attendant Complaints   \n",
       "8769                1.0000                 Lost Luggage   \n",
       "\n",
       "      negative_reason_confidence   airline_name  \\\n",
       "8810                      0.6702  Qatar Airways   \n",
       "9127                      0.6718  Qatar Airways   \n",
       "8771                      0.6902  Qatar Airways   \n",
       "8777                      0.0000  Qatar Airways   \n",
       "9138                      0.6596  Qatar Airways   \n",
       "9150                      0.6632  Qatar Airways   \n",
       "9199                      1.0000  Qatar Airways   \n",
       "9198                      0.3481  Qatar Airways   \n",
       "9202                      1.0000  Qatar Airways   \n",
       "9204                      0.6775  Qatar Airways   \n",
       "9205                      0.6315  Qatar Airways   \n",
       "8795                      0.3681  Qatar Airways   \n",
       "8794                      0.6805  Qatar Airways   \n",
       "8791                      1.0000  Qatar Airways   \n",
       "8805                      1.0000  Qatar Airways   \n",
       "8809                      1.0000  Qatar Airways   \n",
       "8745                      0.3561  Qatar Airways   \n",
       "8783                      0.6736  Qatar Airways   \n",
       "8769                      0.6723  Qatar Airways   \n",
       "\n",
       "                                            review_text  thumbup_count  \\\n",
       "8810  I get some bs from a guy at the door says door...              0   \n",
       "9127  always nice when a customer service manager th...              0   \n",
       "8771  the exhaustion of traveling n runnin thru the ...              0   \n",
       "8777  doesn't take into account the $ it cost rent a...              0   \n",
       "9138  why the hell did you overbook a plane when you...              0   \n",
       "9150  we don't control your stupidity in not having ...              0   \n",
       "9199  this is the reason I avoid and like the plague...              0   \n",
       "9198  how can a plane scheduled to come in on time a...              0   \n",
       "9202  hire smarter IT people if your systems keep cr...              0   \n",
       "9204  what is wrong with you guys and your inability...              1   \n",
       "9205  you guys have to be the worst. We are flying i...              0   \n",
       "8795  you idiots kept an on time plane on the Tarmac...              0   \n",
       "8794  if you actually cared about what you did it wo...              0   \n",
       "8791  is okay for u Cancelled Flight change a flight...              0   \n",
       "8805  just an FYI I'll be writing registered letters...              0   \n",
       "8809  I find it funny that responds but you don't. Y...              0   \n",
       "8745  PLANS CHANGED! IS THAT WHAT U CALL NOT BEIN AB...              0   \n",
       "8783  we had a friend there telling the gate attenda...              0   \n",
       "8769  only happened because u couldn't get us home t...              0   \n",
       "\n",
       "     review_coordinates review_timestamp            review_city user_timezone  \\\n",
       "8810                NaN  23/2/2015 12:38  Does it really matter           NaN   \n",
       "9127                NaN  22/2/2015 19:04  Does it really matter           NaN   \n",
       "8771                NaN  23/2/2015 13:26  Does it really matter           NaN   \n",
       "8777                NaN  23/2/2015 13:22  Does it really matter           NaN   \n",
       "9138                NaN  22/2/2015 18:34  Does it really matter           NaN   \n",
       "9150                NaN  22/2/2015 18:17  Does it really matter           NaN   \n",
       "9199                NaN  22/2/2015 16:49  Does it really matter           NaN   \n",
       "9198                NaN  22/2/2015 16:52  Does it really matter           NaN   \n",
       "9202                NaN  22/2/2015 16:44  Does it really matter           NaN   \n",
       "9204                NaN  22/2/2015 16:42  Does it really matter           NaN   \n",
       "9205                NaN  22/2/2015 16:41  Does it really matter           NaN   \n",
       "8795                NaN  23/2/2015 13:10  Does it really matter           NaN   \n",
       "8794                NaN  23/2/2015 13:11  Does it really matter           NaN   \n",
       "8791                NaN  23/2/2015 13:14  Does it really matter           NaN   \n",
       "8805                NaN  23/2/2015 12:44  Does it really matter           NaN   \n",
       "8809                NaN  23/2/2015 12:40  Does it really matter           NaN   \n",
       "8745                NaN  23/2/2015 14:20  Does it really matter           NaN   \n",
       "8783                NaN  23/2/2015 13:19  Does it really matter           NaN   \n",
       "8769                NaN  23/2/2015 13:29  Does it really matter           NaN   \n",
       "\n",
       "                           review  \n",
       "8810  Flight Attendant Complaints  \n",
       "9127       Customer Service Issue  \n",
       "8771                   Can't Tell  \n",
       "8777                      neutral  \n",
       "9138                   Bad Flight  \n",
       "9150                  Late Flight  \n",
       "9199                   Can't Tell  \n",
       "9198                   Bad Flight  \n",
       "9202  Flight Attendant Complaints  \n",
       "9204                  Late Flight  \n",
       "9205                   Can't Tell  \n",
       "8795                  Late Flight  \n",
       "8794       Customer Service Issue  \n",
       "8791             Cancelled Flight  \n",
       "8805       Customer Service Issue  \n",
       "8809       Customer Service Issue  \n",
       "8745       Customer Service Issue  \n",
       "8783  Flight Attendant Complaints  \n",
       "8769                 Lost Luggage  "
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train['review_city']=='Does it really matter']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "4b4cc6ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "user_to_city_mapping = train[train['user_timezone'] != ''].groupby('user_name')['user_timezone'].first()\n",
    "\n",
    "# Fill in the empty review_city values using the mapping\n",
    "train['user_timezone'] = train.apply(\n",
    "    lambda row: user_to_city_mapping.get(row['user_name'], row['user_timezone']),\n",
    "    axis=1\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "1532667d",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "      <th>review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8336</th>\n",
       "      <td>567776348828360707</td>\n",
       "      <td>somekidnamedjon</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.6485</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>I would love for you to fly my best friend hom...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17/2/2015 12:03</td>\n",
       "      <td>✈️✈️</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               review_id        user_name airline_sentiment  \\\n",
       "8336  567776348828360707  somekidnamedjon          positive   \n",
       "\n",
       "      sentiment_confidence negative_reason  negative_reason_confidence  \\\n",
       "8336                0.6485             NaN                         NaN   \n",
       "\n",
       "            airline_name                                        review_text  \\\n",
       "8336  Singapore Airlines  I would love for you to fly my best friend hom...   \n",
       "\n",
       "      thumbup_count review_coordinates review_timestamp review_city  \\\n",
       "8336              0                NaN  17/2/2015 12:03        ✈️✈️   \n",
       "\n",
       "                   user_timezone    review  \n",
       "8336  Eastern Time (US & Canada)  positive  "
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train['user_name']==\"somekidnamedjon\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92bb064f",
   "metadata": {},
   "source": [
    "we need to identify spams and elimination of data that attribute nothing to the sentimente analysis:<br>\n",
    "we found that news and user spams are both included in the database. \n",
    "In the mean time we also don't see much values in the timezone. As user time zone does not impact sentiment of their \n",
    "Therefore, we decide to not include that column too. \n",
    "\n",
    "We would also get ride off the he thumb_up data as it contains a 94% zero value."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "233adbd3",
   "metadata": {},
   "source": [
    "creat a column that is a count of occurance of the user name  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "65a557b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "train['user_name_count'] = train.groupby('user_name')['user_name'].transform('count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "0fe42342",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "user_name_count = train['user_name_count'].value_counts()\n",
    "\n",
    "# Create a pie chart\n",
    "plt.figure(figsize=(6, 6))\n",
    "plt.pie(user_name_count, labels=user_name_count.index, autopct='%1.1f%%', startangle=90)\n",
    "plt.title('User Name Count Distribution')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a948503",
   "metadata": {},
   "source": [
    "I want to see the distribution for airline sentiment where only user name that contained numbers are selected."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "id": "3c9d56c7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Airline Sentiment Distribution')"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sentiment_counts=filtered_data_number['airline_sentiment'].value_counts()\n",
    "plt.figure(figsize=(6, 6))\n",
    "plt.pie(sentiment_counts, labels=sentiment_counts.index, autopct='%1.1f%%', startangle=90)\n",
    "plt.title('Airline Sentiment Distribution')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "id": "47147207",
   "metadata": {},
   "outputs": [],
   "source": [
    "user_name_counts = train['user_name'].value_counts()\n",
    "filtered_df = train[train['user_name'].isin(user_name_counts[user_name_counts<6].index)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "a1c15705",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows where both 'review_timestamp' and 'review_city' are blank: 2473\n"
     ]
    }
   ],
   "source": [
    "count_blank = len(train[(train['user_timezone'].isna() | train['user_timezone'].eq('')) & (train['review_city'].isna() | train['review_city'].eq(''))])\n",
    "\n",
    "print(\"Number of rows where both 'review_timestamp' and 'review_city' are blank:\", count_blank)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "d7065e89",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows where 'user_timezone', 'review_city', and 'review_coordinates' are blank: 2473\n"
     ]
    }
   ],
   "source": [
    "count_blank2 = len(train[\n",
    "    (train['user_timezone'].isna() | train['user_timezone'].eq('')) &\n",
    "    (train['review_city'].isna() | train['review_city'].eq('')) &\n",
    "    (train['review_coordinates'].isna() | train['review_coordinates'].eq(''))\n",
    "])\n",
    "\n",
    "print(\"Number of rows where 'user_timezone', 'review_city', and 'review_coordinates' are blank:\", count_blank)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "ec49e38f",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "debug_selection = len(train[(train['negative_reason_confidence'].isna()) & (train['airline_sentiment'] == 'negative')])\n",
    "print(debug_selection)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "b0d9bc37",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "highconfidence=train[train['sentiment_confidence']<0.4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "69ee67c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "highconfidenceCha=challenge[challenge['sentiment_confidence']<0.4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "61767815",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "      <th>review</th>\n",
       "      <th>user_name_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6467</th>\n",
       "      <td>570063322449616896</td>\n",
       "      <td>DarthVada_R2D2</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6196</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.3152</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 19:31</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8266</th>\n",
       "      <td>567831806944223232</td>\n",
       "      <td>ProfessorpaUL15</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6472</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>[0.0, 0.0]</td>\n",
       "      <td>17/2/2015 15:43</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>392</th>\n",
       "      <td>568101393111891968</td>\n",
       "      <td>PlantFoodDude</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6559</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>EgyptAir</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 9:35</td>\n",
       "      <td>NJ USA</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9855</th>\n",
       "      <td>569245152406953984</td>\n",
       "      <td>kelleynandrew</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6602</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.3318</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 13:19</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13350</th>\n",
       "      <td>569655113477988352</td>\n",
       "      <td>BurkhalterTrvl</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6604</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:28</td>\n",
       "      <td>Southern Wisconsin</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1690</th>\n",
       "      <td>569580648505020416</td>\n",
       "      <td>mbrooke_brooke</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6628</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6628</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 11:33</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9926</th>\n",
       "      <td>569168696083656705</td>\n",
       "      <td>ThatLittleAnt</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6657</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 8:16</td>\n",
       "      <td>New Haven, CT</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7664</th>\n",
       "      <td>568649053711380480</td>\n",
       "      <td>mhoffma8</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6701</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19/2/2015 21:51</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>neutral</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5205</th>\n",
       "      <td>568911099740839936</td>\n",
       "      <td>AllenSpencer5</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6805</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.6805</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 15:12</td>\n",
       "      <td>Highlands Ranch, CO</td>\n",
       "      <td>Mountain Time (US &amp; Canada)</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5058</th>\n",
       "      <td>569155666440355840</td>\n",
       "      <td>jaindrops</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6842</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6842</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 7:24</td>\n",
       "      <td>Boston, MA</td>\n",
       "      <td>None</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3496</th>\n",
       "      <td>568167900919382017</td>\n",
       "      <td>trueceltsfan</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6871</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.3552</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 13:59</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9930</th>\n",
       "      <td>569165517614944256</td>\n",
       "      <td>lumpuck</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.6910</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.3721</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 8:03</td>\n",
       "      <td>Wexford, PA</td>\n",
       "      <td>None</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8801</th>\n",
       "      <td>569962709057298432</td>\n",
       "      <td>ShawnHFoster</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.6989</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 12:51</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7504</th>\n",
       "      <td>568905143883849728</td>\n",
       "      <td>AroundTownGabby</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.7065</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 14:48</td>\n",
       "      <td>Suburb of Boston, MA</td>\n",
       "      <td>None</td>\n",
       "      <td>positive</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6322</th>\n",
       "      <td>567717985092395008</td>\n",
       "      <td>RachFee</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.9633</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17/2/2015 8:11</td>\n",
       "      <td>London, Ontario, Canada</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9161</th>\n",
       "      <td>569678632567881728</td>\n",
       "      <td>Mark_Shires</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.3515</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 18:02</td>\n",
       "      <td>WITSEC League</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8985</th>\n",
       "      <td>569870445710671872</td>\n",
       "      <td>LarrySandeen</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6760</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>[0.0, 0.0]</td>\n",
       "      <td>23/2/2015 6:44</td>\n",
       "      <td>Southeastern Pennsylvania USA</td>\n",
       "      <td>None</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9110</th>\n",
       "      <td>569703468736376832</td>\n",
       "      <td>Mark_Shires</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6624</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 19:41</td>\n",
       "      <td>WITSEC League</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9104</th>\n",
       "      <td>569708507961950209</td>\n",
       "      <td>PDQuigley</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>0.7160</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 20:01</td>\n",
       "      <td>Washington, D.C.</td>\n",
       "      <td>None</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9549</th>\n",
       "      <td>569507349548957696</td>\n",
       "      <td>HollyAlfano</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6789</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 6:41</td>\n",
       "      <td>Washington, DC</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12541</th>\n",
       "      <td>569884756684476416</td>\n",
       "      <td>61jr</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.6556</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 7:41</td>\n",
       "      <td>St. Catharines</td>\n",
       "      <td>None</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12992</th>\n",
       "      <td>569710119216422912</td>\n",
       "      <td>C2Next</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 20:07</td>\n",
       "      <td>St. Louis, Mo</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12996</th>\n",
       "      <td>569709552721141764</td>\n",
       "      <td>PDQuigley</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Flight Booking Problems</td>\n",
       "      <td>0.3406</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 20:05</td>\n",
       "      <td>Washington, D.C.</td>\n",
       "      <td>None</td>\n",
       "      <td>Flight Booking Problems</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8627</th>\n",
       "      <td>570073507465469952</td>\n",
       "      <td>worldwideweg</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6813</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 20:11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8589</th>\n",
       "      <td>570104544622981122</td>\n",
       "      <td>worldwideweg</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 22:14</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8655</th>\n",
       "      <td>570051488875020288</td>\n",
       "      <td>m2gris</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 18:44</td>\n",
       "      <td>Irvine, Ca</td>\n",
       "      <td>None</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12076</th>\n",
       "      <td>570000777768345600</td>\n",
       "      <td>nf14jk</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 15:22</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10410</th>\n",
       "      <td>568524186185961473</td>\n",
       "      <td>DerekIAm</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19/2/2015 13:35</td>\n",
       "      <td>Miami, FL</td>\n",
       "      <td>Quito</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10970</th>\n",
       "      <td>567832250110337024</td>\n",
       "      <td>mttdprkr</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17/2/2015 15:45</td>\n",
       "      <td>Vancouver, WA</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13362</th>\n",
       "      <td>569653423836045312</td>\n",
       "      <td>dibs65</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qantas</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 16:22</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Cancelled Flight</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9583</th>\n",
       "      <td>569488002361249792</td>\n",
       "      <td>AbiBarrow</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6477</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 5:24</td>\n",
       "      <td>Cambridge, Ma</td>\n",
       "      <td>None</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10204</th>\n",
       "      <td>568812675175591936</td>\n",
       "      <td>MelaniePanton</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 8:41</td>\n",
       "      <td>New York, New York</td>\n",
       "      <td>None</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10144</th>\n",
       "      <td>568861899435016192</td>\n",
       "      <td>SkyDiverChad</td>\n",
       "      <td>positive</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Qatar Airways</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 11:57</td>\n",
       "      <td>Phoenix, AZ</td>\n",
       "      <td>Arizona</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2777</th>\n",
       "      <td>568814123540221952</td>\n",
       "      <td>LifeofPL</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6317</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>[25.0776918, 55.13824067]</td>\n",
       "      <td>20/2/2015 8:47</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1975</th>\n",
       "      <td>569368621077692417</td>\n",
       "      <td>apopnj01</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 21:30</td>\n",
       "      <td>USA</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1977</th>\n",
       "      <td>569368286128771072</td>\n",
       "      <td>21stCenturyMom</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6735</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 21:29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512</th>\n",
       "      <td>570283571941924864</td>\n",
       "      <td>collegepathSS</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>0.6806</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24/2/2015 10:06</td>\n",
       "      <td>California/Nevada</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1577</th>\n",
       "      <td>569629019412557826</td>\n",
       "      <td>RichGoldstein</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>0.7080</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 14:45</td>\n",
       "      <td>New York</td>\n",
       "      <td>None</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585</th>\n",
       "      <td>569626655527981058</td>\n",
       "      <td>nbhdsheila</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 14:35</td>\n",
       "      <td>Boston</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1388</th>\n",
       "      <td>569706336268611584</td>\n",
       "      <td>rlgeraets</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22/2/2015 19:52</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6466</th>\n",
       "      <td>570066166930145280</td>\n",
       "      <td>DarthVada_R2D2</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Damaged Luggage</td>\n",
       "      <td>0.3587</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 19:42</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>Damaged Luggage</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6469</th>\n",
       "      <td>570060184908910592</td>\n",
       "      <td>DarthVada_R2D2</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Lost Luggage</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 19:18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>Lost Luggage</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6282</th>\n",
       "      <td>567734115899027457</td>\n",
       "      <td>laurenproberts</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17/2/2015 9:15</td>\n",
       "      <td>Nashville, TN</td>\n",
       "      <td>None</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7309</th>\n",
       "      <td>569258404742500353</td>\n",
       "      <td>MichaelMontero</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 14:12</td>\n",
       "      <td>Brooklyn, New York</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4870</th>\n",
       "      <td>569273409252827137</td>\n",
       "      <td>nealmo</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Lost Luggage</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 15:12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>Lost Luggage</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4784</th>\n",
       "      <td>569348678911021056</td>\n",
       "      <td>relaxnsmile</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 20:11</td>\n",
       "      <td>Nashville, TN</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>Customer Service Issue</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4041</th>\n",
       "      <td>567716403117957120</td>\n",
       "      <td>tcfitz3</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17/2/2015 8:05</td>\n",
       "      <td>Houston</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>Bad Flight</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4351</th>\n",
       "      <td>569970865607090176</td>\n",
       "      <td>CravenMorhead</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>0.3506</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:23</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mountain Time (US &amp; Canada)</td>\n",
       "      <td>Can't Tell</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5478</th>\n",
       "      <td>568562728585064449</td>\n",
       "      <td>bjpoppe</td>\n",
       "      <td>positive</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19/2/2015 16:08</td>\n",
       "      <td>Columbus, Ohio</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5148</th>\n",
       "      <td>568955169183903745</td>\n",
       "      <td>joecontaldo</td>\n",
       "      <td>negative</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>0.6804</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>#NAME?</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20/2/2015 18:07</td>\n",
       "      <td>Chicago</td>\n",
       "      <td>Mountain Time (US &amp; Canada)</td>\n",
       "      <td>Late Flight</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id        user_name airline_sentiment  \\\n",
       "6467   570063322449616896   DarthVada_R2D2          negative   \n",
       "8266   567831806944223232  ProfessorpaUL15           neutral   \n",
       "392    568101393111891968    PlantFoodDude           neutral   \n",
       "9855   569245152406953984    kelleynandrew          negative   \n",
       "13350  569655113477988352   BurkhalterTrvl           neutral   \n",
       "1690   569580648505020416   mbrooke_brooke          negative   \n",
       "9926   569168696083656705    ThatLittleAnt           neutral   \n",
       "7664   568649053711380480         mhoffma8           neutral   \n",
       "5205   568911099740839936    AllenSpencer5          negative   \n",
       "5058   569155666440355840        jaindrops          negative   \n",
       "3496   568167900919382017     trueceltsfan          negative   \n",
       "9930   569165517614944256          lumpuck          negative   \n",
       "8801   569962709057298432     ShawnHFoster          positive   \n",
       "7504   568905143883849728  AroundTownGabby          positive   \n",
       "6322   567717985092395008          RachFee           neutral   \n",
       "9161   569678632567881728      Mark_Shires          negative   \n",
       "8985   569870445710671872     LarrySandeen          negative   \n",
       "9110   569703468736376832      Mark_Shires          negative   \n",
       "9104   569708507961950209        PDQuigley          negative   \n",
       "9549   569507349548957696      HollyAlfano          negative   \n",
       "12541  569884756684476416             61jr          negative   \n",
       "12992  569710119216422912           C2Next          negative   \n",
       "12996  569709552721141764        PDQuigley          negative   \n",
       "8627   570073507465469952     worldwideweg          negative   \n",
       "8589   570104544622981122     worldwideweg          negative   \n",
       "8655   570051488875020288           m2gris          negative   \n",
       "12076  570000777768345600           nf14jk          negative   \n",
       "10410  568524186185961473         DerekIAm          negative   \n",
       "10970  567832250110337024         mttdprkr          negative   \n",
       "13362  569653423836045312           dibs65          negative   \n",
       "9583   569488002361249792        AbiBarrow          negative   \n",
       "10204  568812675175591936    MelaniePanton          negative   \n",
       "10144  568861899435016192     SkyDiverChad          positive   \n",
       "2777   568814123540221952         LifeofPL          negative   \n",
       "1975   569368621077692417         apopnj01          negative   \n",
       "1977   569368286128771072   21stCenturyMom          negative   \n",
       "512    570283571941924864    collegepathSS          negative   \n",
       "1577   569629019412557826    RichGoldstein          negative   \n",
       "1585   569626655527981058       nbhdsheila           neutral   \n",
       "1388   569706336268611584        rlgeraets          negative   \n",
       "6466   570066166930145280   DarthVada_R2D2          negative   \n",
       "6469   570060184908910592   DarthVada_R2D2          negative   \n",
       "6282   567734115899027457   laurenproberts           neutral   \n",
       "7309   569258404742500353   MichaelMontero          negative   \n",
       "4870   569273409252827137           nealmo          negative   \n",
       "4784   569348678911021056      relaxnsmile          negative   \n",
       "4041   567716403117957120          tcfitz3          negative   \n",
       "4351   569970865607090176    CravenMorhead          negative   \n",
       "5478   568562728585064449          bjpoppe          positive   \n",
       "5148   568955169183903745      joecontaldo          negative   \n",
       "\n",
       "       sentiment_confidence          negative_reason  \\\n",
       "6467                 0.6196               Can't Tell   \n",
       "8266                 0.6472                      NaN   \n",
       "392                  0.6559                      NaN   \n",
       "9855                 0.6602               Can't Tell   \n",
       "13350                0.6604                      NaN   \n",
       "1690                 0.6628              Late Flight   \n",
       "9926                 0.6657                      NaN   \n",
       "7664                 0.6701                      NaN   \n",
       "5205                 0.6805               Can't Tell   \n",
       "5058                 0.6842              Late Flight   \n",
       "3496                 0.6871               Can't Tell   \n",
       "9930                 0.6910               Bad Flight   \n",
       "8801                 0.6989                      NaN   \n",
       "7504                 0.7065                      NaN   \n",
       "6322                 0.9633                      NaN   \n",
       "9161                 1.0000               Can't Tell   \n",
       "8985                 1.0000   Customer Service Issue   \n",
       "9110                 1.0000              Late Flight   \n",
       "9104                 1.0000         Cancelled Flight   \n",
       "9549                 1.0000   Customer Service Issue   \n",
       "12541                1.0000               Can't Tell   \n",
       "12992                1.0000   Customer Service Issue   \n",
       "12996                1.0000  Flight Booking Problems   \n",
       "8627                 1.0000   Customer Service Issue   \n",
       "8589                 1.0000               Can't Tell   \n",
       "8655                 1.0000   Customer Service Issue   \n",
       "12076                1.0000              Late Flight   \n",
       "10410                1.0000   Customer Service Issue   \n",
       "10970                1.0000   Customer Service Issue   \n",
       "13362                1.0000         Cancelled Flight   \n",
       "9583                 1.0000              Late Flight   \n",
       "10204                1.0000               Can't Tell   \n",
       "10144                1.0000                      NaN   \n",
       "2777                 1.0000   Customer Service Issue   \n",
       "1975                 1.0000              Late Flight   \n",
       "1977                 1.0000   Customer Service Issue   \n",
       "512                  1.0000   Customer Service Issue   \n",
       "1577                 1.0000               Bad Flight   \n",
       "1585                 1.0000                      NaN   \n",
       "1388                 1.0000   Customer Service Issue   \n",
       "6466                 1.0000          Damaged Luggage   \n",
       "6469                 1.0000             Lost Luggage   \n",
       "6282                 1.0000                      NaN   \n",
       "7309                 1.0000              Late Flight   \n",
       "4870                 1.0000             Lost Luggage   \n",
       "4784                 1.0000   Customer Service Issue   \n",
       "4041                 1.0000               Bad Flight   \n",
       "4351                 1.0000               Can't Tell   \n",
       "5478                 1.0000                      NaN   \n",
       "5148                 1.0000              Late Flight   \n",
       "\n",
       "       negative_reason_confidence        airline_name review_text  \\\n",
       "6467                       0.3152  Singapore Airlines      #NAME?   \n",
       "8266                       0.0000  Singapore Airlines      #NAME?   \n",
       "392                        0.0000            EgyptAir      #NAME?   \n",
       "9855                       0.3318       Qatar Airways      #NAME?   \n",
       "13350                      0.0000              Qantas      #NAME?   \n",
       "1690                       0.6628            Emirates      #NAME?   \n",
       "9926                          NaN       Qatar Airways      #NAME?   \n",
       "7664                          NaN  Singapore Airlines      #NAME?   \n",
       "5205                       0.6805           Lufthansa      #NAME?   \n",
       "5058                       0.6842           Lufthansa      #NAME?   \n",
       "3496                       0.3552            Emirates      #NAME?   \n",
       "9930                       0.3721       Qatar Airways      #NAME?   \n",
       "8801                       0.0000       Qatar Airways      #NAME?   \n",
       "7504                          NaN  Singapore Airlines      #NAME?   \n",
       "6322                       0.0000           Lufthansa      #NAME?   \n",
       "9161                       0.3515       Qatar Airways      #NAME?   \n",
       "8985                       0.6760       Qatar Airways      #NAME?   \n",
       "9110                       0.6624       Qatar Airways      #NAME?   \n",
       "9104                       0.7160       Qatar Airways      #NAME?   \n",
       "9549                       0.6789       Qatar Airways      #NAME?   \n",
       "12541                      0.6556              Qantas      #NAME?   \n",
       "12992                      1.0000              Qantas      #NAME?   \n",
       "12996                      0.3406              Qantas      #NAME?   \n",
       "8627                       0.6813       Qatar Airways      #NAME?   \n",
       "8589                       1.0000       Qatar Airways      #NAME?   \n",
       "8655                       1.0000       Qatar Airways      #NAME?   \n",
       "12076                      1.0000              Qantas      #NAME?   \n",
       "10410                      1.0000       Qatar Airways      #NAME?   \n",
       "10970                      1.0000       Qatar Airways      #NAME?   \n",
       "13362                      1.0000              Qantas      #NAME?   \n",
       "9583                       0.6477       Qatar Airways      #NAME?   \n",
       "10204                      1.0000       Qatar Airways      #NAME?   \n",
       "10144                         NaN       Qatar Airways      #NAME?   \n",
       "2777                       0.6317            Emirates      #NAME?   \n",
       "1975                       1.0000            Emirates      #NAME?   \n",
       "1977                       0.6735            Emirates      #NAME?   \n",
       "512                        0.6806            Emirates      #NAME?   \n",
       "1577                       0.7080            Emirates      #NAME?   \n",
       "1585                          NaN            Emirates      #NAME?   \n",
       "1388                       1.0000            Emirates      #NAME?   \n",
       "6466                       0.3587  Singapore Airlines      #NAME?   \n",
       "6469                       1.0000  Singapore Airlines      #NAME?   \n",
       "6282                          NaN           Lufthansa      #NAME?   \n",
       "7309                       0.6667  Singapore Airlines      #NAME?   \n",
       "4870                       1.0000           Lufthansa      #NAME?   \n",
       "4784                       1.0000           Lufthansa      #NAME?   \n",
       "4041                       1.0000            Emirates      #NAME?   \n",
       "4351                       0.3506           Lufthansa      #NAME?   \n",
       "5478                          NaN           Lufthansa      #NAME?   \n",
       "5148                       0.6804           Lufthansa      #NAME?   \n",
       "\n",
       "       thumbup_count         review_coordinates review_timestamp  \\\n",
       "6467               0                        NaN  23/2/2015 19:31   \n",
       "8266               0                 [0.0, 0.0]  17/2/2015 15:43   \n",
       "392                0                        NaN   18/2/2015 9:35   \n",
       "9855               0                        NaN  21/2/2015 13:19   \n",
       "13350              0                        NaN  22/2/2015 16:28   \n",
       "1690               0                        NaN  22/2/2015 11:33   \n",
       "9926               0                        NaN   21/2/2015 8:16   \n",
       "7664               0                        NaN  19/2/2015 21:51   \n",
       "5205               1                        NaN  20/2/2015 15:12   \n",
       "5058               0                        NaN   21/2/2015 7:24   \n",
       "3496               0                        NaN  18/2/2015 13:59   \n",
       "9930               0                        NaN   21/2/2015 8:03   \n",
       "8801               0                        NaN  23/2/2015 12:51   \n",
       "7504               0                        NaN  20/2/2015 14:48   \n",
       "6322               0                        NaN   17/2/2015 8:11   \n",
       "9161               0                        NaN  22/2/2015 18:02   \n",
       "8985               0                 [0.0, 0.0]   23/2/2015 6:44   \n",
       "9110               0                        NaN  22/2/2015 19:41   \n",
       "9104               0                        NaN  22/2/2015 20:01   \n",
       "9549               0                        NaN   22/2/2015 6:41   \n",
       "12541              0                        NaN   23/2/2015 7:41   \n",
       "12992              1                        NaN  22/2/2015 20:07   \n",
       "12996              0                        NaN  22/2/2015 20:05   \n",
       "8627               0                        NaN  23/2/2015 20:11   \n",
       "8589               0                        NaN  23/2/2015 22:14   \n",
       "8655               0                        NaN  23/2/2015 18:44   \n",
       "12076              0                        NaN  23/2/2015 15:22   \n",
       "10410              0                        NaN  19/2/2015 13:35   \n",
       "10970              0                        NaN  17/2/2015 15:45   \n",
       "13362              0                        NaN  22/2/2015 16:22   \n",
       "9583               0                        NaN   22/2/2015 5:24   \n",
       "10204              0                        NaN   20/2/2015 8:41   \n",
       "10144              0                        NaN  20/2/2015 11:57   \n",
       "2777               0  [25.0776918, 55.13824067]   20/2/2015 8:47   \n",
       "1975               0                        NaN  21/2/2015 21:30   \n",
       "1977               0                        NaN  21/2/2015 21:29   \n",
       "512                0                        NaN  24/2/2015 10:06   \n",
       "1577               0                        NaN  22/2/2015 14:45   \n",
       "1585               0                        NaN  22/2/2015 14:35   \n",
       "1388               0                        NaN  22/2/2015 19:52   \n",
       "6466               0                        NaN  23/2/2015 19:42   \n",
       "6469               0                        NaN  23/2/2015 19:18   \n",
       "6282               0                        NaN   17/2/2015 9:15   \n",
       "7309               0                        NaN  21/2/2015 14:12   \n",
       "4870               0                        NaN  21/2/2015 15:12   \n",
       "4784               0                        NaN  21/2/2015 20:11   \n",
       "4041               0                        NaN   17/2/2015 8:05   \n",
       "4351               0                        NaN  23/2/2015 13:23   \n",
       "5478               0                        NaN  19/2/2015 16:08   \n",
       "5148               0                        NaN  20/2/2015 18:07   \n",
       "\n",
       "                         review_city                user_timezone  \\\n",
       "6467                             NaN       Atlantic Time (Canada)   \n",
       "8266                             NaN       Atlantic Time (Canada)   \n",
       "392                           NJ USA   Eastern Time (US & Canada)   \n",
       "9855                             NaN                         None   \n",
       "13350             Southern Wisconsin   Central Time (US & Canada)   \n",
       "1690                             NaN                         None   \n",
       "9926                   New Haven, CT   Eastern Time (US & Canada)   \n",
       "7664                             NaN                         None   \n",
       "5205             Highlands Ranch, CO  Mountain Time (US & Canada)   \n",
       "5058                      Boston, MA                         None   \n",
       "3496                             NaN                         None   \n",
       "9930                     Wexford, PA                         None   \n",
       "8801                             NaN   Central Time (US & Canada)   \n",
       "7504            Suburb of Boston, MA                         None   \n",
       "6322         London, Ontario, Canada   Central Time (US & Canada)   \n",
       "9161                   WITSEC League       Atlantic Time (Canada)   \n",
       "8985   Southeastern Pennsylvania USA                         None   \n",
       "9110                   WITSEC League       Atlantic Time (Canada)   \n",
       "9104                Washington, D.C.                         None   \n",
       "9549                  Washington, DC   Eastern Time (US & Canada)   \n",
       "12541                 St. Catharines                         None   \n",
       "12992                  St. Louis, Mo   Central Time (US & Canada)   \n",
       "12996               Washington, D.C.                         None   \n",
       "8627                             NaN                         None   \n",
       "8589                             NaN                         None   \n",
       "8655                      Irvine, Ca                         None   \n",
       "12076                            NaN                         None   \n",
       "10410                      Miami, FL                        Quito   \n",
       "10970                  Vancouver, WA   Pacific Time (US & Canada)   \n",
       "13362                            NaN                         None   \n",
       "9583                   Cambridge, Ma                         None   \n",
       "10204             New York, New York                         None   \n",
       "10144                    Phoenix, AZ                      Arizona   \n",
       "2777                             NaN                         None   \n",
       "1975                             USA   Eastern Time (US & Canada)   \n",
       "1977                             NaN   Pacific Time (US & Canada)   \n",
       "512                California/Nevada   Pacific Time (US & Canada)   \n",
       "1577                        New York                         None   \n",
       "1585                          Boston   Eastern Time (US & Canada)   \n",
       "1388                             NaN                         None   \n",
       "6466                             NaN       Atlantic Time (Canada)   \n",
       "6469                             NaN       Atlantic Time (Canada)   \n",
       "6282                   Nashville, TN                         None   \n",
       "7309              Brooklyn, New York   Eastern Time (US & Canada)   \n",
       "4870                             NaN   Central Time (US & Canada)   \n",
       "4784                   Nashville, TN   Central Time (US & Canada)   \n",
       "4041                         Houston   Central Time (US & Canada)   \n",
       "4351                             NaN  Mountain Time (US & Canada)   \n",
       "5478                  Columbus, Ohio   Eastern Time (US & Canada)   \n",
       "5148                         Chicago  Mountain Time (US & Canada)   \n",
       "\n",
       "                        review  user_name_count  \n",
       "6467                Can't Tell                4  \n",
       "8266                   neutral                2  \n",
       "392                    neutral                1  \n",
       "9855                Can't Tell                2  \n",
       "13350                  neutral                2  \n",
       "1690               Late Flight                3  \n",
       "9926                   neutral                2  \n",
       "7664                   neutral                4  \n",
       "5205                Can't Tell                1  \n",
       "5058               Late Flight                3  \n",
       "3496                Can't Tell                1  \n",
       "9930                Bad Flight                1  \n",
       "8801                  positive                1  \n",
       "7504                  positive                2  \n",
       "6322                   neutral                2  \n",
       "9161                Can't Tell                3  \n",
       "8985    Customer Service Issue                8  \n",
       "9110               Late Flight                3  \n",
       "9104          Cancelled Flight                2  \n",
       "9549    Customer Service Issue                1  \n",
       "12541               Can't Tell                4  \n",
       "12992   Customer Service Issue                2  \n",
       "12996  Flight Booking Problems                2  \n",
       "8627    Customer Service Issue               16  \n",
       "8589                Can't Tell               16  \n",
       "8655    Customer Service Issue                2  \n",
       "12076              Late Flight                4  \n",
       "10410   Customer Service Issue                2  \n",
       "10970   Customer Service Issue                8  \n",
       "13362         Cancelled Flight                2  \n",
       "9583               Late Flight                1  \n",
       "10204               Can't Tell                4  \n",
       "10144                 positive                1  \n",
       "2777    Customer Service Issue                4  \n",
       "1975               Late Flight                3  \n",
       "1977    Customer Service Issue                4  \n",
       "512     Customer Service Issue                1  \n",
       "1577                Bad Flight                2  \n",
       "1585                   neutral                2  \n",
       "1388    Customer Service Issue                1  \n",
       "6466           Damaged Luggage                4  \n",
       "6469              Lost Luggage                4  \n",
       "6282                   neutral                1  \n",
       "7309               Late Flight                1  \n",
       "4870              Lost Luggage                2  \n",
       "4784    Customer Service Issue                1  \n",
       "4041                Bad Flight                3  \n",
       "4351                Can't Tell                3  \n",
       "5478                  positive                1  \n",
       "5148               Late Flight                1  "
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train['review_text']=='#NAME?']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "0ffff27f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ccd3f8691b724d13adccb99ea8debbe5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "                                             |                                             | [  0%]   00:00 ->…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report highconfidence.html was generated! NOTEBOOK/COLAB USERS: the web browser MAY not pop up, regardless, the report IS saved in your notebook/colab files.\n"
     ]
    }
   ],
   "source": [
    "analyze_highconfidence = sv.analyze(highconfidence)\n",
    "analyze_highconfidence.show_html('highconfidence.html')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d8d9ba6",
   "metadata": {},
   "source": [
    "as we can see, the reivew text for sentiments contain things like #Name and thanks. It would be resasonable for us to exclude those from our anlysis as"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e8043abf",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8cb657cb",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f70fab66",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "acf4de12",
   "metadata": {},
   "source": [
    "From the above EDA extracts we can see: <br>\n",
    "1. The airline sentiment are separated into 3 different kinds with 63% in negative, 21% in neural, and 16% in positive.<br>\n",
    "2. There are 10 negative reasons, and the top 1 accounts for 32% being: Customer service Issue.<br>\n",
    "3. We can see the data is collected from 6 different airline companies, with the top 3 being Emirate (26%), Qatar Airways(20%), and Qantas (19%)<br>\n",
    "\n",
    "<br>\n",
    "The data is suffering from missing data:<br>\n",
    "\n",
    "1. 92.24% of review coordinates are missing, we this varible is going to cause collinearity issue with review city and we decide to keep review city. <br>\n",
    "\n",
    "2. Contrary to our believe,review city and user time zone do not share all its missing rows. Only 2499 rows are missing for both review city and review time_zone. The inclusion of review coordinate can bring this down to 2418.<br>\n",
    "\n",
    "3. The missing values of negative_reason is caused by the airline_sentiment not being negative. we are going to create a new column called reivew and backfill \" positive\" or \"neural\" for the missing values.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01354da8",
   "metadata": {},
   "source": [
    "Due to the frequent apperance of unformted and incorrect entries, we are unable to use review city and decide to remove that one.\n",
    "Due to the huge amount of missing data for the review coordianate. we decide to remove this column too."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67fff02f",
   "metadata": {},
   "source": [
    "## Feature Engineering "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "70830b56",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['negative' 'neutral' 'positive']\n"
     ]
    }
   ],
   "source": [
    "# preprocessing data\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "le = LabelEncoder() \n",
    "target = le.fit_transform(train['airline_sentiment']) # convert target into integers\n",
    "train['airline_sentiment'] = target\n",
    "print(le.classes_) # this shows which index maps to which class\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "7d66306f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2724, 15)\n",
      "(2724, 15)\n",
      "(8172, 15)\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(511365)\n",
    "\n",
    "index = list(range(train.shape[0])) # an list of indices\n",
    "np.random.shuffle(index) # shuffle the index in-place\n",
    "\n",
    "p_val = 0.2\n",
    "p_test = 0.2\n",
    "N_test = int(train.shape[0] * p_test)\n",
    "N_val = int(train.shape[0] * p_val)\n",
    "\n",
    "\n",
    "# get training, val and test sets\n",
    "test_data = train.iloc[ index[:N_test] ,:]\n",
    "val_data = train.iloc[ index[N_test: (N_test+N_val)], :]\n",
    "train_data = train.iloc[ index[(N_test+N_val):], :]\n",
    "\n",
    "print(test_data.shape)\n",
    "print(val_data.shape)\n",
    "print(train_data.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "id": "215f18c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define our own torch dataset\n",
    "# for a torch dataset, we need to define two functions: \n",
    "#     __len__: return the length of dataset\n",
    "#     __getitem__: given a index (integer), return the corresponding sample, both y and X\n",
    "\n",
    "class SpamDataset(utils.data.Dataset):\n",
    "    def __init__(self, myData):\n",
    "        \"\"\"\n",
    "        myData should be a dataframe object containing both y (first col) and X (second col)\n",
    "        \"\"\"\n",
    "        super().__init__()\n",
    "        self.data = myData\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.data)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        \n",
    "        return (self.data.iloc[idx,2], self.data.iloc[idx,7]) # (target, text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "6d9af68b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# now we can build our torch dataset \n",
    "train_torch = SpamDataset(train_data)\n",
    "val_torch = SpamDataset(val_data)\n",
    "test_torch = SpamDataset(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "id": "44be9801",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 'where do I look for cabin crew vacancies?')"
      ]
     },
     "execution_count": 183,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check\n",
    "train_torch.__getitem__(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "985e22ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = get_tokenizer('basic_english')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "id": "21585bc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchtext.vocab import build_vocab_from_iterator\n",
    "\n",
    "# ===== Build vocabulary =====\n",
    "# an unknown token is added for all unknown words outside the documents\n",
    "# you may specify the min_freq to filter out infrequent words\n",
    "vocabulary = build_vocab_from_iterator(\n",
    "    [tokenizer(msg) for msg in train['review_text']],\n",
    "    specials=[\"<unk>\"],\n",
    "    min_freq = 3, # filter out all words that appear less than three times\n",
    ")\n",
    "# Set to avoid errors with unknown words\n",
    "vocabulary.set_default_index(vocabulary[\"<unk>\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "id": "45925a7d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4148"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(vocabulary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "id": "4d57a1aa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "224\n",
      "1571\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "# The vocab object maps a word to an idx (an integer)\n",
    "print(vocabulary['better'])\n",
    "print(vocabulary['sun'])\n",
    "print(vocabulary['iertuei']) # something not in vocab will be mapped to default_index = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "id": "a7da99fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define a function that converts a document into tokens (represented by index)\n",
    "def doc_tokenizer(doc):\n",
    "    return torch.tensor([vocabulary[token] for token in tokenizer(doc)], dtype=torch.long)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "id": "97824392",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ========= Step 2 ==============\n",
    "# Notice in a corpus, each document can have different size. Thus, we usually pad zeros to the maximum length of document.\n",
    "# Alternatively, you can concat all documents into a long vector \n",
    "# and the starting index of each document is identified in the variable called offsets.\n",
    "\n",
    "def collate_batch(batch):\n",
    "    \n",
    "    target_list, text_list, offsets = [], [], [0] # initalize the first offset to be 0 here\n",
    "        \n",
    "    # loop through all samples in batch\n",
    "    for idx in range(len(batch)):\n",
    "        \n",
    "        _label = batch[idx][0]\n",
    "        _text = batch[idx][1]\n",
    "        \n",
    "        target_list.append( _label )\n",
    "        tokens = doc_tokenizer( _text )\n",
    "        text_list.append(tokens)\n",
    "        \n",
    "        # ====== wrong ======\n",
    "        #if idx == 0:\n",
    "        #    offsets.append(0)  # the first document starts from idx 0\n",
    "        #else:\n",
    "        #    offsets.append(offsets[-1] + tokens.size(0)) # the next document starts from (offsets[-1] + tokens.size(0))\n",
    "        # ===================\n",
    "        \n",
    "        offsets.append(offsets[-1] + tokens.size(0))\n",
    "    \n",
    "    offsets = offsets[:-1] # remove the last entry\n",
    "    \n",
    "    # convert to torch tensor\n",
    "    target_list = torch.tensor(target_list, dtype=torch.int64)\n",
    "    offsets = torch.tensor(offsets)\n",
    "    text_list = torch.cat(text_list) # concat into a long vector\n",
    "    \n",
    "    return target_list, text_list, offsets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "id": "27c7bfb8",
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.manual_seed(0)\n",
    "\n",
    "batchSize = 8\n",
    "train_loader = utils.data.DataLoader(train_torch, batch_size=batchSize, shuffle=True, collate_fn=collate_batch)\n",
    "val_loader = utils.data.DataLoader(val_torch, batch_size=batchSize, shuffle=True, collate_fn=collate_batch)\n",
    "test_loader = utils.data.DataLoader(test_torch, batch_size=batchSize, shuffle=False, collate_fn=collate_batch)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "id": "dafc74a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([0, 0, 0, 0, 0, 0, 0, 1]),\n",
       " tensor([2582,   15, 1984,   30,  706,  132,    0,  873,   17,   29,    4,  968,\n",
       "           43,    1,  448,  133,   89,    0,  195, 1022,    0, 2213, 1629,    1,\n",
       "          124,   12,  264,  205,   29,  873,  107,    6,  316, 1015,   29,  319,\n",
       "            1,   31,  912,   51,    3,   33, 3209,   81,    2,    0,  914,  142,\n",
       "          359,   18,  253,    3,    5,   59, 1498,    1,  106,   44,   96,  234,\n",
       "           14,   74,   27,  137,    4,   71,    1, 1014, 1002,  105,    9,   14,\n",
       "           22, 2426,   17,    0,    4,  222,   56,  200,    1,   18,    5,   28,\n",
       "           27,    8,  514,   18,    5,   28,    8,    0, 2445,    0,   13,    8,\n",
       "         1392,  202,    1,   16,  421, 3835,  534, 1695,   30,  314,    1,    3,\n",
       "            5,   93,  124,   49, 3484,   38,  172,   13,    8,   12,   29,   15,\n",
       "            0,    1,  368,   11,   15,  554,  113, 1168,  474,   45,    2,   27,\n",
       "          135,  189,   32,   11,    4,  836,  119,   84,  213,   47,    3,  342,\n",
       "           11,  720,  180,    0,    7]),\n",
       " tensor([  0,   6,  32,  56,  81,  99, 122, 140]))"
      ]
     },
     "execution_count": 191,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(train_loader)[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3513cfbb",
   "metadata": {},
   "source": [
    "Model building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "id": "a9eb8dd0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ====== Step 1 ========= \n",
    "class SpamClassifier(nn.Module):\n",
    "    def __init__(self, vocab_size, embed_dim):\n",
    "        super().__init__()\n",
    "        self.embedding = nn.EmbeddingBag(vocab_size, embed_dim, mode='mean') # embedding layer\n",
    "        self.Linear1 = nn.Linear(embed_dim, 1)\n",
    "        self.Dropout = nn.Dropout(p=0.1)\n",
    "    \n",
    "    def forward(self, text, offsets):\n",
    "        # note we need offsets to indicate which document we have\n",
    "        out = self.embedding(text, offsets)\n",
    "        out = self.Dropout(out)\n",
    "        out = self.Linear1(out)\n",
    "        return out\n",
    "        # for the last layer, we don't apply activation because we can use BCEWithLogitsLoss to combine sigmoid with BCELoss\n",
    "        \n",
    "# model initalization\n",
    "embed_dim = 8\n",
    "model = SpamClassifier(len(vocabulary), embed_dim)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "id": "809f7879",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ======= Step 2 ==========\n",
    "loss_fn = torch.nn.BCEWithLogitsLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.005)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "id": "2522214d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate(dataloader):\n",
    "    \n",
    "    y_pred = torch.tensor([]) # store prediction\n",
    "    y_true = torch.tensor([]) # store true label\n",
    "    \n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        for label, text, offsets in dataloader:\n",
    "            y_pred_batch = model(text, offsets)\n",
    "            \n",
    "            y_pred = torch.cat((y_pred, y_pred_batch.squeeze()))\n",
    "            y_true = torch.cat((y_true, label.squeeze()))\n",
    "            \n",
    "    return y_pred, y_true"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "id": "a616c7ca",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: 0.2899 (train), -0.4224 (val), 0.5694 (val acc)\n",
      "Epoch 1: 0.0001 (train), -3.0350 (val), 0.5775 (val acc)\n",
      "Epoch 2: 1.0023 (train), -7.1558 (val), 0.5727 (val acc)\n",
      "Epoch 3: -13.8849 (train), -12.6754 (val), 0.5624 (val acc)\n",
      "Epoch 4: 35.0652 (train), -19.2193 (val), 0.5631 (val acc)\n",
      "Epoch 5: 4.3165 (train), -27.0223 (val), 0.5591 (val acc)\n",
      "Epoch 6: 0.0004 (train), -35.8699 (val), 0.5569 (val acc)\n",
      "Epoch 7: -74.2606 (train), -45.8018 (val), 0.5540 (val acc)\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_23452\\3741001061.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      9\u001b[0m         \u001b[1;31m# calulate output and loss\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     10\u001b[0m         \u001b[0my_pred_train\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtext\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0moffsets\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 11\u001b[1;33m         \u001b[0mloss\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mloss_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my_pred_train\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msqueeze\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfloat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     12\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     13\u001b[0m         \u001b[1;31m# backprop and take a step\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\nn\\modules\\module.py\u001b[0m in \u001b[0;36m_wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_compiled_call_impl\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m  \u001b[1;31m# type: ignore[misc]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1517\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1518\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call_impl\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1519\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1520\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_call_impl\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\nn\\modules\\module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1525\u001b[0m                 \u001b[1;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1526\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1528\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1529\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\nn\\modules\\loss.py\u001b[0m in \u001b[0;36mforward\u001b[1;34m(self, input, target)\u001b[0m\n\u001b[0;32m    723\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    724\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtarget\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 725\u001b[1;33m         return F.binary_cross_entropy_with_logits(input, target,\n\u001b[0m\u001b[0;32m    726\u001b[0m                                                   \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mweight\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    727\u001b[0m                                                   \u001b[0mpos_weight\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpos_weight\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\nn\\functional.py\u001b[0m in \u001b[0;36mbinary_cross_entropy_with_logits\u001b[1;34m(input, target, weight, size_average, reduce, reduction, pos_weight)\u001b[0m\n\u001b[0;32m   3193\u001b[0m         \u001b[1;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf\"Target size ({target.size()}) must be the same as input size ({input.size()})\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3194\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3195\u001b[1;33m     \u001b[1;32mreturn\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbinary_cross_entropy_with_logits\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtarget\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpos_weight\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreduction_enum\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3196\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3197\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# ======== Step 3 ==============\n",
    "epochs = 300\n",
    "for epoch in range(epochs):\n",
    "    \n",
    "    for y_train, text, offsets in train_loader:\n",
    "        # zero the parameter gradients\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        # calulate output and loss \n",
    "        y_pred_train = model(text, offsets)\n",
    "        loss = loss_fn(y_pred_train.squeeze(), y_train.float())\n",
    "\n",
    "        # backprop and take a step\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "    \n",
    "    # evaluate on validation set\n",
    "    y_pred_val, y_val = evaluate(val_loader)\n",
    "    loss_val = loss_fn(y_pred_val.squeeze(), y_val.float())\n",
    "    \n",
    "    # note when making prediction, do add sigmoid activation\n",
    "    pred_label = (torch.sigmoid(y_pred_val) > 0.5).long() # find out the class prediction\n",
    "    acc = (pred_label == y_val).float().sum()/y_val.shape[0]\n",
    "    \n",
    "    model.train() # because when evaluating we change mode to eval mode\n",
    "    \n",
    "    print('Epoch {}: {:.4f} (train), {:.4f} (val), {:.4f} (val acc)'.format(epoch, loss, loss_val, acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e097d95a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# prediction on test data\n",
    "y_pred_test, y_true_test = evaluate(test_loader)\n",
    "y_pred_test = torch.sigmoid(y_pred_test) > 0.5\n",
    "\n",
    "print(confusion_matrix(y_true_test, y_pred_test))\n",
    "print(classification_report(y_true_test, y_pred_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1e768ad",
   "metadata": {},
   "source": [
    "using word embedding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "id": "abeb4e0f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ===== Build vocabulary =====\n",
    "vocab2 = build_vocab_from_iterator(\n",
    "    [tokenizer(s) for s in train],\n",
    "    specials=[\"<unk>\"]\n",
    ")\n",
    "\n",
    "vocab2.set_default_index(vocab2[\"<unk>\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "id": "adfb9b63",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to convert a sentence to BoW sequence\n",
    "vocab2_len = len(vocab2)\n",
    "def index2onehot(idx, n = vocab2_len):\n",
    "    eye = torch.eye(vocab2_len)\n",
    "    return eye[:,idx]\n",
    "\n",
    "def stobow(s):\n",
    "    token_idxs = vocab2(tokenizer(s))\n",
    "    return index2onehot(token_idxs), torch.Tensor(token_idxs).long()\n",
    "\n",
    "# Convert all training sentences\n",
    "bow_sequences = []\n",
    "for s in train['review_text']:\n",
    "    bow_sequences.append(stobow(s)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "818c20e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Model class\n",
    "class RNN(nn.Module):\n",
    "    def __init__(self, n_features, n_class, hidden_dim=20):\n",
    "        super().__init__()\n",
    "        # Model layers\n",
    "        self.rnn = nn.RNN(n_features, hidden_dim)\n",
    "        self.fc = nn.Linear(hidden_dim, n_class)\n",
    "        \n",
    "    def forward(self, x):\n",
    "        rnn_o, h = self.rnn(x)\n",
    "        fc_o = self.fc(rnn_o[-1, :]) \n",
    "        return fc_o\n",
    "\n",
    "# init model\n",
    "model = RNN(n_features=vocab2_len, n_class=vocab2_len)\n",
    "\n",
    "# init loss\n",
    "loss_function = nn.CrossEntropyLoss()\n",
    "\n",
    "# init optimizer\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr = 0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "id": "93e2e177",
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_23452\\821130632.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     13\u001b[0m             \u001b[0my\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtokens\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mseq_length\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;31m# j+2 word as label\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 15\u001b[1;33m             \u001b[0moptimizer\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     16\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     17\u001b[0m             \u001b[0moutput\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mT\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;31m# pytorch requires input to be [Length, inputsize]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\_compile.py\u001b[0m in \u001b[0;36minner\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     22\u001b[0m             \u001b[1;32mimport\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_dynamo\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     23\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 24\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_dynamo\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdisable\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mrecursive\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     25\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     26\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0minner\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\_dynamo\\eval_frame.py\u001b[0m in \u001b[0;36m_fn\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m    326\u001b[0m             \u001b[0mdynamic_ctx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__enter__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    327\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 328\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    329\u001b[0m             \u001b[1;32mfinally\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    330\u001b[0m                 \u001b[0mset_eval_frame\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mprior\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\optim\\optimizer.py\u001b[0m in \u001b[0;36mzero_grad\u001b[1;34m(self, set_to_none)\u001b[0m\n\u001b[0;32m    821\u001b[0m                 \u001b[1;32mfor\u001b[0m \u001b[0mper_dtype_grads\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mper_device_and_dtype_grads\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    822\u001b[0m                     \u001b[1;32mfor\u001b[0m \u001b[0mgrads\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mper_dtype_grads\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 823\u001b[1;33m                         \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_foreach_zero_\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mgrads\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    824\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    825\u001b[0m     \u001b[1;33m@\u001b[0m\u001b[0moverload\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\autograd\\profiler.py\u001b[0m in \u001b[0;36m__exit__\u001b[1;34m(self, exc_type, exc_value, traceback)\u001b[0m\n\u001b[0;32m    646\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mjit\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mis_scripting\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    647\u001b[0m             \u001b[1;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_C\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDisableTorchFunctionSubclass\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 648\u001b[1;33m                 \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprofiler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_record_function_exit\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_RecordFunction\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrecord\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    649\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    650\u001b[0m             \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprofiler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_record_function_exit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrecord\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\Annaconda\\lib\\site-packages\\torch\\_ops.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m    446\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    447\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 448\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_op\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m \u001b[1;32mor\u001b[0m \u001b[1;33m{\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    449\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    450\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m__hash__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "max_iter = 300\n",
    "seq_length = 2\n",
    "loss_history = []\n",
    "for epoch in range(max_iter):\n",
    "    \n",
    "    for i in range(len(bow_sequences)):\n",
    "    \n",
    "        Sent, tokens = bow_sequences[i]\n",
    "        \n",
    "        for j in range(Sent.shape[1]-seq_length):\n",
    "            # loop over the two-grams\n",
    "            X = Sent[:, j:j+seq_length] # j, j+1 words\n",
    "            y = tokens[j+seq_length] # j+2 word as label\n",
    "        \n",
    "            optimizer.zero_grad()\n",
    "\n",
    "            output = model(X.T) # pytorch requires input to be [Length, inputsize]\n",
    "                        \n",
    "            loss = loss_function(output, y)\n",
    "\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "    # Collect loss at end of each iteration\n",
    "    loss_history.append(loss.item())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "30e08d76",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure()\n",
    "plt.plot(loss_history)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47adad15",
   "metadata": {},
   "source": [
    "Spam Classification \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "id": "212e0e95",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>sentiment_confidence</th>\n",
       "      <th>negative_reason</th>\n",
       "      <th>negative_reason_confidence</th>\n",
       "      <th>airline_name</th>\n",
       "      <th>review_text</th>\n",
       "      <th>thumbup_count</th>\n",
       "      <th>review_coordinates</th>\n",
       "      <th>review_timestamp</th>\n",
       "      <th>review_city</th>\n",
       "      <th>user_timezone</th>\n",
       "      <th>review</th>\n",
       "      <th>user_name_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>930</th>\n",
       "      <td>569972097453137920</td>\n",
       "      <td>cristobalwong</td>\n",
       "      <td>1</td>\n",
       "      <td>0.3350</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>Thank you, ^JH, appreciate the prompt response...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23/2/2015 13:28</td>\n",
       "      <td>San Francisco Bay Area</td>\n",
       "      <td>None</td>\n",
       "      <td>neutral</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5873</th>\n",
       "      <td>568092537786748928</td>\n",
       "      <td>VegasRenegade</td>\n",
       "      <td>1</td>\n",
       "      <td>0.3353</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>All Flights lead to VEGAS BABY!</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 8:59</td>\n",
       "      <td>The Wild Blue Yonder</td>\n",
       "      <td>Atlantic Time (Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5963</th>\n",
       "      <td>568028183267639297</td>\n",
       "      <td>CourtSnod</td>\n",
       "      <td>2</td>\n",
       "      <td>0.3356</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Lufthansa</td>\n",
       "      <td>de-icing is important!</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18/2/2015 4:44</td>\n",
       "      <td>Tennessee Original</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7436</th>\n",
       "      <td>568993773277069312</td>\n",
       "      <td>KoolieAshSays</td>\n",
       "      <td>1</td>\n",
       "      <td>0.3363</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Singapore Airlines</td>\n",
       "      <td>oh. Makes sense. My bad.</td>\n",
       "      <td>0</td>\n",
       "      <td>[40.88529232, -73.856635]</td>\n",
       "      <td>20/2/2015 20:41</td>\n",
       "      <td>ÜT: 17.9889591,-76.7712636</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2254</th>\n",
       "      <td>569227372223811584</td>\n",
       "      <td>zachkpearson</td>\n",
       "      <td>2</td>\n",
       "      <td>0.3366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Emirates</td>\n",
       "      <td>. Thanks. Hopefully this is easily resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21/2/2015 12:09</td>\n",
       "      <td>Portland, Oregon</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               review_id      user_name  airline_sentiment  \\\n",
       "930   569972097453137920  cristobalwong                  1   \n",
       "5873  568092537786748928  VegasRenegade                  1   \n",
       "5963  568028183267639297      CourtSnod                  2   \n",
       "7436  568993773277069312  KoolieAshSays                  1   \n",
       "2254  569227372223811584   zachkpearson                  2   \n",
       "\n",
       "      sentiment_confidence negative_reason  negative_reason_confidence  \\\n",
       "930                 0.3350             NaN                         0.0   \n",
       "5873                0.3353             NaN                         0.0   \n",
       "5963                0.3356             NaN                         0.0   \n",
       "7436                0.3363             NaN                         0.0   \n",
       "2254                0.3366             NaN                         0.0   \n",
       "\n",
       "            airline_name                                        review_text  \\\n",
       "930             Emirates  Thank you, ^JH, appreciate the prompt response...   \n",
       "5873           Lufthansa                    All Flights lead to VEGAS BABY!   \n",
       "5963           Lufthansa                             de-icing is important!   \n",
       "7436  Singapore Airlines                           oh. Makes sense. My bad.   \n",
       "2254            Emirates       . Thanks. Hopefully this is easily resolved.   \n",
       "\n",
       "      thumbup_count         review_coordinates review_timestamp  \\\n",
       "930               0                        NaN  23/2/2015 13:28   \n",
       "5873              1                        NaN   18/2/2015 8:59   \n",
       "5963              0                        NaN   18/2/2015 4:44   \n",
       "7436              0  [40.88529232, -73.856635]  20/2/2015 20:41   \n",
       "2254              0                        NaN  21/2/2015 12:09   \n",
       "\n",
       "                     review_city               user_timezone    review  \\\n",
       "930       San Francisco Bay Area                        None   neutral   \n",
       "5873        The Wild Blue Yonder      Atlantic Time (Canada)   neutral   \n",
       "5963          Tennessee Original  Central Time (US & Canada)  positive   \n",
       "7436  ÜT: 17.9889591,-76.7712636  Pacific Time (US & Canada)   neutral   \n",
       "2254            Portland, Oregon  Pacific Time (US & Canada)  positive   \n",
       "\n",
       "      user_name_count  \n",
       "930                 7  \n",
       "5873                1  \n",
       "5963                1  \n",
       "7436                1  \n",
       "2254                1  "
      ]
     },
     "execution_count": 209,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# preprocessing data\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "le = LabelEncoder() \n",
    "target = le.fit_transform(train['airline_sentiment']) # convert target into integers\n",
    "train['airline_sentiment'] = target\n",
    "print(le.classes_) # this shows which index maps to which class\n",
    "\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d498e8d",
   "metadata": {},
   "source": [
    "simplified adaboost (week 11)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}