- A text mining web application which performs sentimental analysis of Enterprises on Twitter Dataset and displays a graphical representation of the different sentiments of the tweets.
- It can also generate a Word Cloud of what social media thinks about a company by analysing the tweets of a company. After processing the tweets it also displays a histogram of words which a related to a particular company and a table of most recent tweets.
Right now the application has been deployed and tested in Windows environment, so the following steps guide you on installing this application on Windows Environment.
- R - Download and install R from R for Windows
- RStudio - Download and install RStudio Desktop from RStudio Desktop for Windows
- XAMPP Stack - Download and install XAMPP from XAMPP for Windows
- Download the zip of this project.
- Extract in xampp\htdocs\
- Open wordcloud.php , update line 26 with your database settings.
- Open wordcloud.R located in bin , update line 43 with your database settings.
- Install the following packages in R - twitteR, wordcloud, tm , ggplot2, plyr.
- The application needs an authentication from Twitter via oauth protocol. I have provided the API keys myself.
- If you want to use your own API keys, go Twitter apps , click on create a new app
- You may have to provide a website name for authentication.
- After succesully creating a new app, go to Keys and Access tokens in application settings
- Copy Consumer Key(API Secret), Consumer Secret(API Secret), Access Token , Access Token Secret.
- Paste the keys in wordcloud.R - line 11-14 and sentimentalAnalysis.R - line 14.
- Now run XAMPP as administrator(very important!)
- Fire up localhost , and you application is up and ready.
- Enter the company names you want to do analysis about.
- It can take a little bit time to process the results depending on your Internet Connection.
- Stay Calm and Composed , Enjoy !!!