Week-11.1_Clustering_solutions.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# *Unsupervised learning: k-means clustering*\n",
    "\n",
    "[http://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html](http://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Importing basic packages\n",
    "\n",
    "import os\n",
    "import random\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "## Downloading several hundred New York Times articles as text files\n",
    "\n",
    "os.chdir('/sharedfolder/')\n",
    "\n",
    "!wget -N https://github.com/pcda17/pcda17.github.io/raw/master/week/10/nyt_articles_11-9-2017.zip\n",
    "!unzip -o nyt_articles_11-9-2017.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "451\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"The seal stone comes from an untouched shaft grave near the ancient palace of Pylos. The grave was discovered in May 2015 by Jack L. Davis and Sharon R. Stocker, archaeologists at the University of Cincinnati who had been digging at Pylos for more than 25 years.  “It was after cleaning, during the process of drawing and photography, that our excitement slowly rose as we gradually came to realize that we had unearthed a masterpiece,” they wrote in the journal Hesperia.  Photo  The seal stone presents two mysteries. One is how and why it was engraved in such detail. The other is whether its battle scene, strongly evocative of those in Homer’s “Iliad” and “Odyssey,” depicts an event that contributed to the oral tradition behind the works of Homer.  The seal stone’s owner, known as the Griffin Warrior after the mythical animal depicted in his grave, was buried around 1450 B.C. He lived at a critical period when the Minoan civilization of Crete was being transferred to cities of the Greek mainland.  Local chieftains, as the Griffin Warrior may have been, used precious items from Crete to advertise their membership in the Greek-speaking elite of the incipient Mycenaean civilization, the first on mainland Europe. Their descendants, a century or so later, built the great palaces at Pylos, Mycenae and Tiryns, places mentioned by Homer.  Dr. Davis and Dr. Stocker believe that the seal stone, like other objects in the Griffin Warrior’s grave, was made on Crete. Work of such quality was not being produced anywhere on the Greek mainland at the time. The detail is so fine that it seems the engraver would have needed a magnifying glass, as would admirers of his work.  Yet no magnifying implements have been found on Crete from this era. Perhaps the engraver was nearsighted, the two archaeologists suggest.  Advertisement Continue reading the main story  Fritz Blakolmer, an expert on Aegean art at the University of Vienna, argues that the seal stone is a miniature copy of a much larger original, probably a stucco-embellished wall painting like those found at the Palace of Knossos on Crete. He said the seal must have been engraved by someone with a magnifying glass, even though none has been found, and dismissed the possibility that people of that era had sharper eyesight than today.  More Reporting on Archaeology  The seal, carved on a hard stone known as agate, shows a victorious hero slaying an adversary while a third warrior lies dead in the foreground. The seal stone is mounted so that it can be worn on the wrist, and indeed the hero is wearing just such an item, as if it were a wristwatch.  The two defeated warriors seem to belong to the same group, because both are wearing patterned kilts whereas the hero sports a codpiece. The scene evidently represents some event that would have been familiar to the Minoans who made the seal stone and to the Griffin Warrior’s community.  Photo  The seal stone’s possible relevance to the Homeric epics is intriguing but elusive. Early archaeologists, such as Heinrich Schliemann, who first excavated Troy and Mycenae, believed the “Iliad” recounted historical events and were quick to see proof of this in the artifacts they found.  Later archaeologists were more doubtful, but allowed that the destruction of Troy in 1200 B.C. could have been remembered in oral poetry for 500 years until the Homeric poems were first written down, around 700 B.C.  The Griffin Warrior was buried around 1450 B.C., distancing him even further from the first written version of Homer. Still, there is some evidence that the oral tradition behind the Homeric epics traces as far back as Linear B, the first Greek writing system.  Newsletter Sign Up Continue reading the main story Please verify you're not a robot by clicking the box. Invalid email address. Please re-enter. You must select a newsletter to subscribe to. Sign Up You agree to receive occasional updates and special offers for The New York Times's products and services. Thank you for subscribing. An error has occurred. Please try again later. View all New York Times newsletters.  Linear B was adapted by the Mycenaean Greeks from Linear A, used by the Minoans. The oldest known Linear B inscriptions date to 1450 B.C., and the script disappeared after the collapse of Mycenaean civilization around 1200 B.C.  Some of the scansion problems in the Homeric poems “can be resolved if you restore older forms of Greek which are consistent with the dialect recorded in Linear B documents,” said Dr. Bennet of the British School at Athens.  So the oral tradition that led to the Homeric epics perhaps stretched over seven centuries.  “We’re not saying this is a representation from Homer,” Dr. Stocker said of the tableau on the seal stone, while admitting it would be “fun to believe” the hero is Achilles. Rather, the image “is part of a cycle of stories familiar to both Mycenaeans and Minoans.”  Dr. Blakolmer, too, finds it tempting to see the figures on the seal as Homeric heroes, like Hector or Nestor, but in his view the temptation must be resisted.  Advertisement Continue reading the main story  “Fifty years ago, you would find nice attributions to Homeric heroes, but today’s academics are very careful in their Homeric attributions,” he said. “We have to make our own mistakes, not theirs.”\""
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Loading all text files in the current directory as a list of strings\n",
    "\n",
    "os.chdir('/sharedfolder/nyt_articles_11-9-2017/')\n",
    "\n",
    "document_list = []\n",
    "\n",
    "for filename in [item for item in os.listdir('./') if '.txt' in item]: # Excluding files other than .txt with a list comprehension\n",
    "    text_data = open(filename).read()\n",
    "    document_list.append(text_data)\n",
    "\n",
    "print(len(document_list)) # Printing number of documents in list\n",
    "\n",
    "random.choice(document_list) # Viewing a single document chosen at random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', 'your']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Creating stop word list\n",
    "\n",
    "from nltk.corpus import stopwords\n",
    "import string\n",
    "\n",
    "stop_words = stopwords.words('english') + [\"'s\", \"'re\", '”', '“', '‘', '’', '—'] + list(string.punctuation)\n",
    "\n",
    "stop_words[:10]     ## Viewing first 10 stop words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Tokenizing, stemming, and removing stop words from our list of documents\n",
    "\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.stem.porter import PorterStemmer\n",
    "\n",
    "stemmer = PorterStemmer()\n",
    "\n",
    "documents_filtered = []\n",
    "\n",
    "for document in document_list:\n",
    "    token_list = word_tokenize(document.lower())                                 ## Tokenizing\n",
    "    tokens_filtered = [item for item in token_list if (item not in stop_words)]  ## Removing stop words\n",
    "    tokens_filtered = [stemmer.stem(item) for item in tokens_filtered]           ## Stemming\n",
    "    documents_filtered.append(' '.join(tokens_filtered))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'blockbust season video game holiday season way game industri push divers lineup titl new instal long-run franchis like activis call duti ubisoft assassin creed ea long-await star war battlefront ii real crowd pleaser year may hardwar microsoft nintendo offer new upgrad consol soni rereleas playstat 4 consol power ps4 pro last year hottest hardwar ticket nintendo switch consol offer gamer option plug home pull dock play go even though releas eight month ago switch still popular suppli expect problem end year arriv week microsoft new xbox one x consol featur process power 4k resolut mean defin pictur qualiti also expect make wave consol support hundr game xbox back catalog includ xbox one titl well select xbox 360 titl advertis continu read main stori expect fuel spend game rose 7 percent 14 billion first half year period 2016 accord npd group expect growth continu edg holiday said mat piscatella analyz video game industri npd'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Viewing a preprocessed document\n",
    "\n",
    "random.choice(documents_filtered)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Vectorizing preprocessed documents\n",
    "\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "\n",
    "vectorizer = CountVectorizer()\n",
    "\n",
    "X = vectorizer.fit_transform(documents_filtered) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['amort',\n",
       " 'amount',\n",
       " 'amped',\n",
       " 'ampl',\n",
       " 'amplif',\n",
       " 'amplifi',\n",
       " 'ampush',\n",
       " 'amsterdam',\n",
       " 'amtrak',\n",
       " 'amulet',\n",
       " 'amus',\n",
       " 'ana',\n",
       " 'anab',\n",
       " 'anal',\n",
       " 'analog',\n",
       " 'analogu',\n",
       " 'analys',\n",
       " 'analysi',\n",
       " 'analyst',\n",
       " 'analyt']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Creating a vocabulary list corresponding to the vectors we created above\n",
    "\n",
    "vocabulary = vectorizer.get_feature_names()\n",
    "\n",
    "vocabulary[1140:1160]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 2, 2, 2, 2, 2, 2, 2, 2, 6, 2, 2, 2, 2, 0, 2, 0, 2, 2, 2], dtype=int32)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Clustering documents\n",
    "\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "kmeans_classifier = KMeans(n_clusters=8, init='k-means++', n_init=10, max_iter=500, algorithm='auto')\n",
    "                                   ## ^ docuements to be grouped in 8 clusters \n",
    "\n",
    "cluster_classes = kmeans_classifier.fit_predict(X) \n",
    "\n",
    "cluster_classes[:20]               ## Viewing first 20 cluster assignments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "451\n",
      "451\n",
      "451\n"
     ]
    }
   ],
   "source": [
    "## Our cluster assignments and document lists are the same size, in the same order.\n",
    "\n",
    "print(len(cluster_classes))\n",
    "\n",
    "print(len(documents_filtered))\n",
    "\n",
    "print(len(document_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cluster assignment:\n",
      "2\n",
      "\n",
      "Document:\n",
      "“The one thing that I never wanted for him was to feel any type of racism where he felt like he was being targeted simply because he’s an African-American male,” Ms. Carr said.  It is unclear who took the video, which appears to have been recorded with a phone, and when. Ms. Carr said her daughter brought it to her attention on Nov. 2.  Ms. Carr asked that her son’s name not be published because he is a minor and “not the type of kid that really likes a lot of attention.” She said he told her that he had been laughing with some friends during a physics lesson, and that the teacher had gotten frustrated.  “I believe that he did like his teacher, and he believed that his teacher liked him,” she said, adding that Mr. Hagan apologized to her son.  Mr. Hagan did not respond to phone calls or an email seeking comment on Wednesday.  Ms. Carr said she contacted the Rockdale County Sheriff’s Department. A spokesman for the department said on Tuesday that a report had been filed, but he would not say whether the department was conducting an investigation of its own, referring the question to the school board.  The Rockdale County school district is investigating the episode, said Cindy Ball, a spokeswoman. “Rockdale County Public Schools expects all employees to conduct themselves professionally and ethically to provide a positive teaching and learning environment for students and staff,” she added in an emailed statement. She confirmed that Mr. Hagan was on paid administrative leave.  Advertisement Continue reading the main story  Ms. Carr is calling for Mr. Hagan to be fired.  “I hate that my son had to be the example,” she said. “But this is about awareness and fairness. We don’t send our kids to school to be threatened. We send them to be educated and protected.”\n"
     ]
    }
   ],
   "source": [
    "## We can view a document and its assigned cluster by accessing the same index in both lists\n",
    "\n",
    "index_num = 130\n",
    "\n",
    "print('Cluster assignment:')\n",
    "print(cluster_classes[index_num])\n",
    "print()\n",
    "print('Document:')\n",
    "print(document_list[index_num])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Write each document to a new text file, with each cluster of documents in its own directory\n",
    "\n",
    "try: os.mkdir('/sharedfolder/nyt_clusters_11-9-2017/')\n",
    "except: pass\n",
    "\n",
    "for i in range(len(documents_filtered)):\n",
    "    \n",
    "    out_dir = '/sharedfolder/nyt_clusters_11-9-2017/Cluster_' + str(cluster_classes[i])  ## Creating a directory pathname that\n",
    "                                                                                         ## includes the assigned cluster number.\n",
    "    try: os.mkdir(out_dir)  ## Creating the out_dir directory if it does not yet exist\n",
    "    except: pass\n",
    "    \n",
    "    os.chdir(out_dir)\n",
    "    \n",
    "    out_filename = 'Document_' + str(i) + '.txt'    ## Creating a filename for the text file\n",
    "    \n",
    "    with open(out_filename, 'w') as file_out:\n",
    "        file_out.write(document_list[i])            ## Writing text from original (non-preprocessed) document list\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2], dtype=int32)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Classifying a new text into an existing cluster\n",
    "\n",
    "input_vector = vectorizer.transform(['Even the budget office is revising its estimates and has predicted the new numbers would be smaller.'])\n",
    "\n",
    "kmeans_classifier.predict(input_vector)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *Scatter Plots*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAGfCAYAAABC5ObhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f77516a1d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## A simple scatter plot example\n",
    "\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 7)  # Setting default plot size\n",
    "\n",
    "x_vals_1 = [1.92, 1.79, 1.96, 1.4, 1.61, 1.23, 1.43, 1.85, 2.07, 2.24, 2.11, 1.78, 2.21, 1.79, 1.33]\n",
    "y_vals_1 = [2.5, 2.11, 2.19, 1.6, 2.83, 2.55, 2.33, 2.09, 2.32, 2.72, 2.05, 2.4, 2.55, 2.83, 1.58]\n",
    "\n",
    "x_vals_2 = [3.63, 3.12, 3.21, 3.15, 3.56, 3.17, 3.05, 3.14, 2.87, 3.65, 2.82, 3.34, 3.7, 2.95, 3.15]\n",
    "y_vals_2 = [3.1, 4.27, 4.03, 3.37, 3.22, 3.89, 3.27, 2.64, 3.09, 4.1, 3.61, 3.74, 3.71, 3.51, 2.9]\n",
    "\n",
    "\n",
    "plt.scatter(x_vals_1, y_vals_1, label='Class 1', c='indigo', alpha=0.5)\n",
    "\n",
    "plt.scatter(x_vals_2, y_vals_2, label='Class 2', c='blue', alpha=0.5)\n",
    "\n",
    "plt.ylim(ymin=0, ymax=5)\n",
    "plt.xlim(xmin=0, xmax=5)\n",
    "\n",
    "plt.legend(loc=4)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.5/dist-packages/sklearn/discriminant_analysis.py:442: UserWarning: The priors do not sum to 1. Renormalizing\n",
      "  UserWarning)\n",
      "/usr/local/lib/python3.5/dist-packages/sklearn/discriminant_analysis.py:388: UserWarning: Variables are collinear.\n",
      "  warnings.warn(\"Variables are collinear.\")\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlwAAAGfCAYAAACKiAw3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f774b0e5208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Projecting all vectors to 2 dimensions using linear discriminant analysis (LDA) and viewing a scatter plot\n",
    "\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 7)  # Setting default plot size\n",
    "\n",
    "lda = LDA(n_components=2) #2-dimensional LDA\n",
    "\n",
    "y = cluster_classes\n",
    "\n",
    "lda_transformed = pd.DataFrame(lda.fit_transform(X.toarray(), y))\n",
    "\n",
    "lda_transformed['y'] = y\n",
    "\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==0][0], lda_transformed[lda_transformed['y']==0][1], label='0', c='blue', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==1][0], lda_transformed[lda_transformed['y']==1][1], label='1', c='green', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==2][0], lda_transformed[lda_transformed['y']==2][1], label='2', c='red', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==3][0], lda_transformed[lda_transformed['y']==3][1], label='3', c='violet', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==4][0], lda_transformed[lda_transformed['y']==4][1], label='4', c='orange', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==5][0], lda_transformed[lda_transformed['y']==5][1], label='5', c='magenta', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==6][0], lda_transformed[lda_transformed['y']==6][1], label='6', c='black', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==7][0], lda_transformed[lda_transformed['y']==7][1], label='7', c='indigo', alpha=0.4)\n",
    "\n",
    "plt.legend(loc=2)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlsAAAGfCAYAAABoVBdOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f77512f03c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Projecting all vectors to 2 dimensions using principal component analysis (PCA) and viewing a scatter plot\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 7)  # Setting default plot size\n",
    "\n",
    "pca = PCA(n_components=2) #2-dimensional LDA\n",
    "\n",
    "pca_transformed = pca.fit_transform(X.toarray())\n",
    "\n",
    "plt.scatter([pair[0] for pair in pca_transformed], [pair[1] for pair in pca_transformed], label='0', c='blue', alpha=0.4)\n",
    "\n",
    "plt.legend(loc=2)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### *Clustering in topic space*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "!pip3 install lda"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:lda:n_documents: 451\n",
      "INFO:lda:vocab_size: 18084\n",
      "INFO:lda:n_words: 215533\n",
      "INFO:lda:n_topics: 8\n",
      "INFO:lda:n_iter: 1000\n",
      "INFO:lda:<0> log likelihood: -2393892\n",
      "INFO:lda:<10> log likelihood: -2014323\n",
      "INFO:lda:<20> log likelihood: -1943768\n",
      "INFO:lda:<30> log likelihood: -1916179\n",
      "INFO:lda:<40> log likelihood: -1903004\n",
      "INFO:lda:<50> log likelihood: -1895432\n",
      "INFO:lda:<60> log likelihood: -1890202\n",
      "INFO:lda:<70> log likelihood: -1886172\n",
      "INFO:lda:<80> log likelihood: -1882790\n",
      "INFO:lda:<90> log likelihood: -1881253\n",
      "INFO:lda:<100> log likelihood: -1878386\n",
      "INFO:lda:<110> log likelihood: -1877507\n",
      "INFO:lda:<120> log likelihood: -1876716\n",
      "INFO:lda:<130> log likelihood: -1875726\n",
      "INFO:lda:<140> log likelihood: -1874009\n",
      "INFO:lda:<150> log likelihood: -1873489\n",
      "INFO:lda:<160> log likelihood: -1872991\n",
      "INFO:lda:<170> log likelihood: -1871191\n",
      "INFO:lda:<180> log likelihood: -1870671\n",
      "INFO:lda:<190> log likelihood: -1871127\n",
      "INFO:lda:<200> log likelihood: -1870061\n",
      "INFO:lda:<210> log likelihood: -1870073\n",
      "INFO:lda:<220> log likelihood: -1868546\n",
      "INFO:lda:<230> log likelihood: -1869260\n",
      "INFO:lda:<240> log likelihood: -1868797\n",
      "INFO:lda:<250> log likelihood: -1867713\n",
      "INFO:lda:<260> log likelihood: -1867267\n",
      "INFO:lda:<270> log likelihood: -1867609\n",
      "INFO:lda:<280> log likelihood: -1866628\n",
      "INFO:lda:<290> log likelihood: -1866893\n",
      "INFO:lda:<300> log likelihood: -1866330\n",
      "INFO:lda:<310> log likelihood: -1866128\n",
      "INFO:lda:<320> log likelihood: -1865501\n",
      "INFO:lda:<330> log likelihood: -1864469\n",
      "INFO:lda:<340> log likelihood: -1864504\n",
      "INFO:lda:<350> log likelihood: -1864123\n",
      "INFO:lda:<360> log likelihood: -1865171\n",
      "INFO:lda:<370> log likelihood: -1864338\n",
      "INFO:lda:<380> log likelihood: -1864510\n",
      "INFO:lda:<390> log likelihood: -1864194\n",
      "INFO:lda:<400> log likelihood: -1864684\n",
      "INFO:lda:<410> log likelihood: -1864747\n",
      "INFO:lda:<420> log likelihood: -1863900\n",
      "INFO:lda:<430> log likelihood: -1863741\n",
      "INFO:lda:<440> log likelihood: -1863783\n",
      "INFO:lda:<450> log likelihood: -1863224\n",
      "INFO:lda:<460> log likelihood: -1863785\n",
      "INFO:lda:<470> log likelihood: -1863952\n",
      "INFO:lda:<480> log likelihood: -1863270\n",
      "INFO:lda:<490> log likelihood: -1861825\n",
      "INFO:lda:<500> log likelihood: -1861608\n",
      "INFO:lda:<510> log likelihood: -1861503\n",
      "INFO:lda:<520> log likelihood: -1861849\n",
      "INFO:lda:<530> log likelihood: -1861532\n",
      "INFO:lda:<540> log likelihood: -1860782\n",
      "INFO:lda:<550> log likelihood: -1860889\n",
      "INFO:lda:<560> log likelihood: -1861049\n",
      "INFO:lda:<570> log likelihood: -1859974\n",
      "INFO:lda:<580> log likelihood: -1859662\n",
      "INFO:lda:<590> log likelihood: -1859316\n",
      "INFO:lda:<600> log likelihood: -1858667\n",
      "INFO:lda:<610> log likelihood: -1858596\n",
      "INFO:lda:<620> log likelihood: -1857721\n",
      "INFO:lda:<630> log likelihood: -1857975\n",
      "INFO:lda:<640> log likelihood: -1857778\n",
      "INFO:lda:<650> log likelihood: -1857872\n",
      "INFO:lda:<660> log likelihood: -1856831\n",
      "INFO:lda:<670> log likelihood: -1857965\n",
      "INFO:lda:<680> log likelihood: -1857244\n",
      "INFO:lda:<690> log likelihood: -1858114\n",
      "INFO:lda:<700> log likelihood: -1858114\n",
      "INFO:lda:<710> log likelihood: -1857324\n",
      "INFO:lda:<720> log likelihood: -1857593\n",
      "INFO:lda:<730> log likelihood: -1857209\n",
      "INFO:lda:<740> log likelihood: -1856411\n",
      "INFO:lda:<750> log likelihood: -1856702\n",
      "INFO:lda:<760> log likelihood: -1856216\n",
      "INFO:lda:<770> log likelihood: -1855825\n",
      "INFO:lda:<780> log likelihood: -1855751\n",
      "INFO:lda:<790> log likelihood: -1855872\n",
      "INFO:lda:<800> log likelihood: -1855810\n",
      "INFO:lda:<810> log likelihood: -1855826\n",
      "INFO:lda:<820> log likelihood: -1854771\n",
      "INFO:lda:<830> log likelihood: -1855077\n",
      "INFO:lda:<840> log likelihood: -1855310\n",
      "INFO:lda:<850> log likelihood: -1855564\n",
      "INFO:lda:<860> log likelihood: -1855199\n",
      "INFO:lda:<870> log likelihood: -1854780\n",
      "INFO:lda:<880> log likelihood: -1854396\n",
      "INFO:lda:<890> log likelihood: -1853942\n",
      "INFO:lda:<900> log likelihood: -1853807\n",
      "INFO:lda:<910> log likelihood: -1854071\n",
      "INFO:lda:<920> log likelihood: -1854336\n",
      "INFO:lda:<930> log likelihood: -1853786\n",
      "INFO:lda:<940> log likelihood: -1853378\n",
      "INFO:lda:<950> log likelihood: -1852638\n",
      "INFO:lda:<960> log likelihood: -1852731\n",
      "INFO:lda:<970> log likelihood: -1851410\n",
      "INFO:lda:<980> log likelihood: -1851858\n",
      "INFO:lda:<990> log likelihood: -1850982\n",
      "INFO:lda:<999> log likelihood: -1851069\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<lda.lda.LDA at 0x7f7746a13198>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import lda\n",
    "\n",
    "model = lda.LDA(n_topics=8, n_iter=1000, random_state=1)\n",
    "\n",
    "model.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "451\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[  9.73025048e-02,   2.12909441e-01,   2.12909441e-01,\n",
       "          9.63391137e-04,   9.63391137e-04,   9.63391137e-04,\n",
       "          9.63391137e-04,   4.73025048e-01],\n",
       "       [  6.56669463e-01,   8.82969799e-02,   2.11828859e-02,\n",
       "          2.09731544e-04,   1.90855705e-02,   3.79614094e-02,\n",
       "          1.76384228e-01,   2.09731544e-04]])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_features = model.doc_topic_\n",
    "\n",
    "print(len(topic_features))\n",
    "\n",
    "topic_features[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "451\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([1, 7, 6, 4, 2, 6, 5, 7, 7, 1, 7, 0, 3, 5, 3, 5, 2, 5, 0, 3], dtype=int32)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Clustering documents using topic vectors (instead of vocabulary vectors, which we used above)\n",
    "\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "kmeans_topic_classifier = KMeans(n_clusters=8, init='k-means++', n_init=15, max_iter=1000, algorithm='auto')\n",
    "                                         ## ^ docuements to be grouped in 8 clusters \n",
    "\n",
    "topic_cluster_classes = kmeans_topic_classifier.fit_predict(topic_features) \n",
    "\n",
    "print(len(topic_cluster_classes))\n",
    "\n",
    "topic_cluster_classes[:20]               ## Viewing first 20 cluster assignments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Write each document to a new text file, with each cluster of documents in its own directory\n",
    "\n",
    "try: os.mkdir('/sharedfolder/nyt_topic_clusters_11-9-2017/')\n",
    "except: pass\n",
    "\n",
    "for i in range(len(documents_filtered)):\n",
    "    \n",
    "    out_dir = '/sharedfolder/nyt_topic_clusters_11-9-2017/Cluster_' + str(topic_cluster_classes[i])  ## Creating a directory pathname that\n",
    "                                                                                         ## includes the assigned cluster number.\n",
    "    try: os.mkdir(out_dir)  ## Creating the out_dir directory if it does not yet exist\n",
    "    except: pass\n",
    "    \n",
    "    os.chdir(out_dir)\n",
    "    \n",
    "    out_filename = 'Document_' + str(i) + '.txt'    ## Creating a filename for the text file\n",
    "    \n",
    "    with open(out_filename, 'w') as file_out:\n",
    "        file_out.write(document_list[i])            ## Writing text from original (non-preprocessed) document list\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.5/dist-packages/sklearn/discriminant_analysis.py:388: UserWarning: Variables are collinear.\n",
      "  warnings.warn(\"Variables are collinear.\")\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlUAAAGfCAYAAAB2nSf9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCCGEKIL/H1y0M+YtHxisAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f777478e550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Projecting all topic vectors to 2 dimensions using linear discriminant analysis (LDA) and viewing a scatter plot\n",
    "\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 7)  # Setting default plot size\n",
    "\n",
    "lda = LDA(n_components=2) #2-dimensional LDA\n",
    "\n",
    "y = topic_cluster_classes\n",
    "\n",
    "lda_transformed = pd.DataFrame(lda.fit_transform(topic_features, y))\n",
    "\n",
    "lda_transformed['y'] = y\n",
    "\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==0][0], lda_transformed[lda_transformed['y']==0][1], label='0', c='blue', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==1][0], lda_transformed[lda_transformed['y']==1][1], label='1', c='green', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==2][0], lda_transformed[lda_transformed['y']==2][1], label='2', c='red', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==3][0], lda_transformed[lda_transformed['y']==3][1], label='3', c='violet', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==4][0], lda_transformed[lda_transformed['y']==4][1], label='4', c='yellow', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==5][0], lda_transformed[lda_transformed['y']==5][1], label='5', c='magenta', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==6][0], lda_transformed[lda_transformed['y']==6][1], label='6', c='black', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==7][0], lda_transformed[lda_transformed['y']==7][1], label='7', c='orange', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==8][0], lda_transformed[lda_transformed['y']==8][1], label='8', c='indigo', alpha=0.4)\n",
    "plt.scatter(lda_transformed[lda_transformed['y']==9][0], lda_transformed[lda_transformed['y']==9][1], label='9', c='purple', alpha=0.4)\n",
    "\n",
    "plt.legend(loc=2)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Topic cluster: 0\n",
      "[('said', 329),\n",
      " ('mr.', 296),\n",
      " ('new', 277),\n",
      " ('time', 245),\n",
      " ('ms.', 212),\n",
      " ('york', 198),\n",
      " ('like', 190),\n",
      " ('one', 184),\n",
      " ('stori', 134),\n",
      " ('imag', 124),\n",
      " ('year', 121),\n",
      " ('photo', 117),\n",
      " ('newslett', 117),\n",
      " ('work', 108),\n",
      " ('pleas', 106),\n",
      " ('read', 104),\n",
      " ('credit', 101),\n",
      " ('also', 101),\n",
      " ('first', 101),\n",
      " ('look', 95),\n",
      " ('continu', 95),\n",
      " ('show', 94),\n",
      " ('main', 89),\n",
      " ('say', 85),\n",
      " ('peopl', 83)]\n",
      "\n",
      "Topic cluster: 1\n",
      "[('mr.', 629),\n",
      " ('said', 424),\n",
      " ('trump', 297),\n",
      " ('new', 293),\n",
      " ('time', 265),\n",
      " ('democrat', 264),\n",
      " ('state', 260),\n",
      " ('main', 208),\n",
      " ('would', 207),\n",
      " ('republican', 203),\n",
      " ('continu', 202),\n",
      " ('newslett', 198),\n",
      " ('stori', 196),\n",
      " ('read', 190),\n",
      " ('pleas', 184),\n",
      " ('presid', 180),\n",
      " ('york', 179),\n",
      " ('year', 172),\n",
      " ('sign', 171),\n",
      " ('one', 156),\n",
      " ('elect', 155),\n",
      " ('like', 149),\n",
      " ('also', 137),\n",
      " ('peopl', 135),\n",
      " ('parti', 127)]\n",
      "\n",
      "Topic cluster: 2\n",
      "[('car', 382),\n",
      " ('compani', 236),\n",
      " ('said', 195),\n",
      " ('like', 187),\n",
      " ('new', 178),\n",
      " ('would', 149),\n",
      " ('one', 145),\n",
      " ('could', 143),\n",
      " ('time', 133),\n",
      " ('vehicl', 132),\n",
      " ('year', 128),\n",
      " ('tesla', 122),\n",
      " ('use', 114),\n",
      " ('make', 114),\n",
      " ('ad', 112),\n",
      " ('self-driv', 109),\n",
      " ('technolog', 108),\n",
      " ('facebook', 99),\n",
      " ('also', 96),\n",
      " ('peopl', 96),\n",
      " ('say', 88),\n",
      " ('get', 88),\n",
      " ('autonom', 87),\n",
      " ('driver', 81),\n",
      " ('go', 80)]\n",
      "\n",
      "Topic cluster: 3\n",
      "[('time', 205),\n",
      " ('new', 187),\n",
      " ('servic', 175),\n",
      " ('use', 167),\n",
      " ('said', 164),\n",
      " ('inform', 152),\n",
      " ('tax', 152),\n",
      " ('may', 138),\n",
      " ('york', 127),\n",
      " ('state', 108),\n",
      " ('pleas', 99),\n",
      " ('colleg', 97),\n",
      " ('would', 93),\n",
      " ('includ', 92),\n",
      " ('email', 84),\n",
      " ('year', 84),\n",
      " ('chang', 82),\n",
      " ('student', 81),\n",
      " ('percent', 76),\n",
      " ('continu', 75),\n",
      " ('newslett', 75),\n",
      " ('get', 74),\n",
      " ('product', 73),\n",
      " ('one', 68),\n",
      " ('read', 68)]\n",
      "\n",
      "Topic cluster: 4\n",
      "[('said', 258),\n",
      " ('like', 234),\n",
      " ('time', 226),\n",
      " ('one', 179),\n",
      " ('stori', 168),\n",
      " ('new', 163),\n",
      " ('continu', 154),\n",
      " ('read', 153),\n",
      " ('newslett', 142),\n",
      " ('main', 133),\n",
      " ('pleas', 131),\n",
      " ('would', 130),\n",
      " ('mr.', 124),\n",
      " ('get', 121),\n",
      " ('tri', 116),\n",
      " ('go', 111),\n",
      " ('say', 110),\n",
      " ('peopl', 110),\n",
      " ('york', 109),\n",
      " ('know', 106),\n",
      " ('year', 105),\n",
      " ('want', 105),\n",
      " ('sign', 101),\n",
      " ('work', 100),\n",
      " ('famili', 93)]\n",
      "\n",
      "Topic cluster: 5\n",
      "[('new', 231),\n",
      " ('time', 187),\n",
      " ('york', 177),\n",
      " ('newslett', 174),\n",
      " ('pleas', 153),\n",
      " ('said', 137),\n",
      " ('sign', 117),\n",
      " ('subscrib', 107),\n",
      " ('mr.', 106),\n",
      " ('read', 93),\n",
      " ('continu', 90),\n",
      " ('stori', 88),\n",
      " ('main', 82),\n",
      " ('email', 69),\n",
      " ('updat', 66),\n",
      " ('would', 64),\n",
      " ('like', 64),\n",
      " ('offer', 62),\n",
      " ('servic', 62),\n",
      " ('view', 61),\n",
      " ('tri', 60),\n",
      " ('later', 56),\n",
      " ('special', 56),\n",
      " ('must', 56),\n",
      " ('one', 55)]\n",
      "\n",
      "Topic cluster: 6\n",
      "[('said', 168),\n",
      " ('new', 152),\n",
      " ('time', 133),\n",
      " ('like', 116),\n",
      " ('one', 112),\n",
      " ('read', 107),\n",
      " ('continu', 105),\n",
      " ('stori', 104),\n",
      " ('main', 98),\n",
      " ('newslett', 96),\n",
      " ('pleas', 93),\n",
      " ('team', 82),\n",
      " ('york', 82),\n",
      " ('wine', 80),\n",
      " ('sign', 78),\n",
      " ('game', 78),\n",
      " ('year', 76),\n",
      " ('make', 75),\n",
      " ('tri', 75),\n",
      " ('also', 71),\n",
      " ('advertis', 71),\n",
      " ('player', 68),\n",
      " ('would', 66),\n",
      " ('food', 66),\n",
      " ('season', 61)]\n",
      "\n",
      "Topic cluster: 7\n",
      "[('said', 183),\n",
      " ('new', 149),\n",
      " ('park', 101),\n",
      " ('read', 91),\n",
      " ('main', 91),\n",
      " ('time', 91),\n",
      " ('one', 90),\n",
      " ('year', 86),\n",
      " ('newslett', 86),\n",
      " ('ms.', 82),\n",
      " ('york', 82),\n",
      " ('stori', 81),\n",
      " ('continu', 77),\n",
      " ('hous', 73),\n",
      " ('sign', 72),\n",
      " ('island', 70),\n",
      " ('street', 70),\n",
      " ('pleas', 70),\n",
      " ('mr.', 70),\n",
      " ('includ', 68),\n",
      " ('like', 67),\n",
      " ('home', 66),\n",
      " ('price', 66),\n",
      " ('photo', 65),\n",
      " ('percent', 63)]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## Viewing the most frequently used words in each topic cluster\n",
    "\n",
    "import collections\n",
    "from pprint import pprint\n",
    "\n",
    "\n",
    "for topic_cluster_id in range(8):\n",
    "    \n",
    "    print('Topic cluster: ' + str(topic_cluster_id))    \n",
    "    \n",
    "    words = []\n",
    "    \n",
    "    for i in range(len(documents_filtered)):\n",
    "        if topic_cluster_classes[i] == topic_cluster_id:\n",
    "            words += documents_filtered[i].split(' ')\n",
    "\n",
    "    counter = collections.Counter(words)\n",
    "    pprint(counter.most_common()[:25])\n",
    "    print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}