Deciphering Personality Disorders in the Digital Realm: A Deep Learning Approach Based on LSTM to Social Media Analysis

This paper examines the intersection of mental health and technology, specifically focusing on the detection of personality disorders through social media data and deep learning techniques. Building on foundational research, we analyze a diverse dataset of 27,977 individuals. Our findings underscore the potential of linguistic patterns combined with deep learning to identify indicators of personality disorders. This study contributes to the ethical discourse on the use of technology in mental health, highlighting the importance of advancing diagnostic tools and interventions. Our results are comparable to state-of-the-art methods, achieving an accuracy of 0.89 and a recall of 0.95.