Predicting National Self-Harm Trends via Social Network Analysis

The project uses Bi-LSTM-RNN to predict self-harm severity using social media data, integrating engagement metrics, sentiment analysis, and demographics. It preprocesses diverse social media datasets for optimal model performance, leveraging the architecture's ability to handle sequential data effectively. Initial findings show high accuracy in predicting self-harm risks,intervention by healthcare professionals and policymakers. This approach enhances understanding and prevention strategies for self-harm, impacting mental health services and public safety initiatives significantly.