In this project the tweets are extracted manually using tweets.py and collected over 20k+ tweets from twitter using different keyword into one CSV file (fina_output.csv). Tweets till 5000 are labelled manually and used as a dataset for this project.
Procedure to run through the whole project:
- Run tweets.py to collect tweet from the twitter.
- label them as depressed or not (can also use the database i used)
- Run the main.py file to run the model