Project Info:

For this hackathon, our group built a NLP model to diagnose mental illnesses. It takes in a piece of text, whether that be a sentence or a paragraph, and diagnoses mental illnesses based on that information. Our model is currently able to detect 9 different mental illnesses. We used BERT and fine-tuned the classification layer to match our classes. Data was taken from an open-source scrape of the major mental health subreddits. This project won the best healthcare hack!

Project Details:

• Tools: Python, Tensorflow, AWS, Pandas
• Date:January 2022
• URL:Devpost