Steps followed-
- Scrape data from Reddit.
- Clean the data.
- Preprocess the data for EDA.
- EDA (Histograms, WordClouds).
- Build and analyze ML models.
- Prepare data for BERT model.
- Build, train and test the BERT model.
Refer to the PDF for report and output analysis. Repeat the confusion matrix building step (cell) for all the models.