A Multi-Modal Approach for Predicting Mental Health Risk using Social Media Data

This thesis explores how artificial intelligence can detect early signs of depression from social media posts. It develops a deep learning model that analyzes both text and images to identify patterns linked to depression. The model is tested across different social media platforms to ensure it works well beyond a single platform.This research helps create better tools for early mental health screening, allowing professionals to identify at-risk individuals sooner and provide timely support.