Sentiment Prediction and graphing of live twitter data with recurrent nets and lstm

This project analyses live twitter sentiments and visualize them using recurrent neural networks and long short term memories. I have used recurrent nets because while training on huge data, recurrent nets actually predict the outcome a lot better than any normal machine learning models.

*** This project may throw errors if trained on CPU instead of GPU ***

This project is configured for Linux and uses python3

To run this project, open up your bash terminal and write

chmod -R 777 setup.sh
./setup.sh

This will set up the project enviornment for you. This must be run with administrator rights. After you set up the project, run:

Setup Virtual enviornment

virtualenv -p python3 venv
source venv/bin/activate
pip install -r requirements.txt

Train and test the accuracy of the project

python configure.py

Analyse and visualise twitter data

python run_me.py

keyword = Ransomware

Plots:

keyword = Modi

keyword = Trump

keyword = Ransomware

Author: Satyaki Sanyal

*** This project is strictly for educational purposes only. ***