Predicting Digital Addiction Patterns with Machine Learning for Personalised Mental Health Support

Project Code :TCMAPY2228

Objective

The primary objective of this project is to develop a machine learning framework to predict digital addiction patterns, leveraging structured behavioral data from individuals. This framework aims to assess digital addiction risk through a comprehensive analysis of factors such as screen time, social media engagement, sleep duration, and mental health indicators. By employing various machine learning models, including Random Forest Regressor, Ridge Regressor, and Support Vector Regression (SVR), the project seeks to identify key predictors of digital addiction. Furthermore, the project aims to provide personalized, data-driven recommendations for mental health support through clustering and predictive modeling techniques. The overarching goal is to enhance the understanding of digital addiction, guide non-invasive intervention strategies, and inform scalable solutions to improve mental health outcomes.

Abstract

The project titled "Predicting Digital Addiction Patterns with Machine Learning for Personalised Mental Health Support" aims to predict the mental health status of individuals based on various lifestyle and technology usage factors. The dataset includes features such as age, technology usage hours, social media usage hours, gaming hours, screen time hours, sleep hours, physical activity hours, gender, stress level, support systems access, work environment impact, and online support usage. The system uses machine learning algorithms like SVM, Random Forest, and a hybrid model combining CatBoost and GNN for predictions. The backend of the system is built using Django, and the frontend uses HTML, CSS, and JavaScript. The project is designed to offer personalized mental health support by analyzing digital addiction patterns and providing predictive insights into mental well-being.