/NLP-Depression-Analysis

Depression analysis and prediction using posts/comments scraped from Reddit.

Primary LanguageJupyter NotebookMIT LicenseMIT

Depression-Analysis-NLP

Steps followed-

  1. Scrape data from Reddit.
  2. Clean the data.
  3. Preprocess the data for EDA.
  4. EDA (Histograms, WordClouds).
  5. Build and analyze ML models.
  6. Prepare data for BERT model.
  7. 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.