This paper presents a pioneering approach leveraging machine learning algorithms to detect mental health issues from online social media interactions. This paper delves into the fusion of computational techniques and psychological insights, analyzing user-generated content and behavioral patterns to identify potential indicators of mental disorders. By harnessing the vast data repository of social media, this research aims to develop proactive screening methods, facilitating early detection and intervention for individuals at risk. The study underscores the promise of technology in augmenting mental health support systems, paving the way for more accessible and timely assistance in the digital landscape. Intellectual disorders prediction is based on predictions from the Random Forest and Naive Bayes algorithms. It is proved that Naive Bayes performs better than Random Forest algorithm in terms of accuracy.

Machine Learning Algorithms, Online Social Media Interactions, Mental Health Issues Detection, User-Generated Content Analysis, Proactive Screening Methods, Naive Bayes, Random Forest Algorithms