A Systematic Review of Machine Learning Algorithms for Mental Health Detection Using Social Media Data

Abstract

Mental health plays a key role in our daily lives, affecting our thoughts, behaviors, and relationships. However, mental health issues like depression, anxiety, and stress are becoming more common due to multiple factors such as family problems, failed relationships, and social demands. Early detection and timely treatment are crucial to prevent harm to individuals and society. This systematic review looks at how machine learning can help identify mental health issues by analyzing data from social media. We review methods used in recent studies, such as natural language processing, sentiment analysis, and boosting algorithms, based on 100 peer-reviewed articles. These methods are grouped into supervised machine learning and unsupervised machine learning algorithms. Finally, we identify gaps such as a lack of clinical assessment and psychological acknowledgment in current studies. Overall, we suggest integrating psychological acknowledgment with machine learning to improve accuracy and responsibility in using machine learning for mental health assessment on social media.