indranil143/Mental-Health-Sentiment-Analysis-using-Deep-Learning
A deep learning project using fine-tuned RoBERTa to classify mental health sentiments from text, aiming to provide early insights and support. ⚕️❤️
Fine-tunes RoBERTa with balanced class weights and AdamW optimization to classify text into seven mental health categories (Anxiety, Depression, Suicidal, etc.), achieving 75.33% accuracy over a baseline Logistic Regression model. The pipeline includes comprehensive text preprocessing with contraction expansion, stopword removal, and lemmatization, alongside EDA with n-gram and word cloud analysis across 50,000+ labeled samples from Kaggle.
No commits in the last 6 months.
Stars
35
Forks
7
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jul 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/indranil143/Mental-Health-Sentiment-Analysis-using-Deep-Learning"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
galihru/MentalHealth
A comprehensive mental health monitoring application using modern web technologies.
giastahmad/DSS-Anxiety-Evaluator
Web app for anxiety assessment using DASS-42 and an ML model to predict key contributing factors.
mistr3ated/AI-Psychometrics-Nigel
This repository contains code from my experiments with AI for psychological measurement. Read...
zeinhasan/Suicidal-Detection-Sentiment-Analysis
Suicidal Ideation Detection Using Natural Languange Processing and Machine Learning - Deep...
BouzidiImen/Social_media_Prediction_depression
This consists in using a variety of social networks data, including both images and texts, to...