Skip to content

A mini-project which performs sentiment analysis on tweets and displays it as a nice GUI.

Notifications You must be signed in to change notification settings

Kumar-Tarun/analyze360

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

51 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

analyze360 is a web app(a python flask app) based on sentimental analysis. It searches for tweets containing the topic typed in by the user and performs the required analysis on it.

It has the following features-
   1)Overall sentiment level
   2)No.of tweets sentiment wise and day wise
   3)Latest tweet
   4)Geospatial plotting of the tweets(country wise).

It uses the module tweepy for the twitter API and the classifier used is Naive-Bayes Classifier.
The required python modules are contained in requirements.txt.

gettweet.py gets the tweets based on user's searches.
application.py is the crux of the website backend where the real analysis of the tweets happens.
Procfile is a file required by heroku to deploy websites.

You need to install NLTK corpora using following command-
    python -m textblob.download_corpora

The app also uses analyze.dump which contains tables for different countries, timezones, latitudes and longitudes which is used for the plotting of tweets on the world map.

The website is hosted on heroku on the following URL - https://analyze360.herokuapp.com/