Detecting Mental Disorders in Social Media Through Emotional Patterns: The Case of Anorexia and Depression

The rise of social media as a primary channel for communication has enabled researchers to explore the digital footprints of users for signs of psychological wellbeing or distress.This study focuses on detecting mental disorders-specifically anorexia and depression-through the analysis of emotional patterns in social media posts.By using natural language processing (NLP) and machine learning techniques, emotional cues such as tone, sentiment, frequency, and engagement are extracted from online user-generated content.The goal is to identify linguistic and behavioral signals indicative of mental health struggles.The proposed model achieves effective classification by analyzing emotional shifts and semantic content over time, offering a non-invasive and scalable solution for early detection.Results indicate that social media can serve as a valuable tool for monitoring public mental health and potentially intervening before clinical symptoms become severe.