Analysis of Social Media Impacts on Mental Health using machine learning

Technological advancement has made social media a central part of our lives, significantly affecting various parts of our mental well-being. In spite of it offering multiple opportunities for social interaction and providing unlimited learning resources, social media also has its share of advantages and disadvantages. Social media makes it possible to raise awareness about natural disasters, political, and social causes. The primary objective of this study is to analyse the relationship between social media usage and mental health using advanced machine learning techniques. This study made use of a publicly available Kaggle dataset with information on the social media usage and mental health data across 100 000 individuals. The study is expected to provide meaningful insights into the relationship between social media usage and mental well-being. The data set is analysed for a result that produces a better understanding of the correlation between users’ digital habits and their mental health. The data set was cleaned, processed and analysed using Google Colab. After cleaning the data and processing the data, the Random Forest Regressor model was developed to predict mental health scores based on users’ digital habits. The final results showed that the model achieved a Mean Absolute Error of 0.544, a Root Mean Squared Error of 0.809, and 0.596 as the R squared score achieving an averagely strong prediction accuracy. These findings provide a deeper insight of the correlation between social media usage patterns and mental well-being