Flask application hosting Twitter API. Main functionality allows user to find out what is currently trending and extract data sample to MongoDB. Once data sample is extracted, next step is to pre-process it and load to SQL database. Final step is to apply Keras LSTM model on processed tweets to find out what's the dominating sentiment among conversation participants - positive, negative or neutral.
Choose between app main functionalities:
Find out what is currently trending in US. Pick most interesting subject and guide your data sample through ETL pipeline to make it ready for sentiment analysis.
Get rid of obsolete collections from MongoDB or delete processed data from SQLite database:
Apply Deep learning LSTM model on previously stored Twitter dataset. Find out what's the dominating sentiment among conversation participants - positive, negative or neutral:
Find out what is currently trending in two chosen locations. Specify woe_id and find out common trends:
Choose a keyword and find out most popular tweets in a subject:
Find out what are the top Twitter trends in US at the moment. Trends are updated every 60 seconds.
Flask | REST | Tweepy | Oauthlib | Pickle
MongoDB | Sqlite
nltk | re
keras | tensorflow | numpy | scikit-learn | h5py
HTML | CSS | Bootstrap | Materialize | Conda | Heroku | Docker
selenium | unittest
Thank you,
Lukasz Malucha