Enhanced Conditional Generative Adversarial Networks Suicidal Risk Identification from Social Networking Sites Using Text Similarity Measures

Abstract

Publicly available social networks can be exploited for monitoring mental healthcare by employing machine learning (ML) and Artificial Intelligence (AI) methods to classify and assess the risk related with different mental health illness. By integrating text similarity measurements with Deep Learning (DL) methods, the CGANSRI model develops the accuracy of risk assessment by efficiently capturing contextual nuances. The CGANSRI can discern subtle linguistic cues that may recommend suicidal thoughts through its training on a vast corpus of social media posts. Concurrently, sophisticated measures of text similarity are employed to quantify the semantic relationships between posts, providing a comprehensive understanding of the textual content. The system efficiently investigates information provided by users to organize nuanced linguistic cues that may be associated with the risk of suicide by incorporating ATSM. This method develops the accuracy and responsiveness of organizing suicidal risk by using the combined benefits of online communications, despite the challenges posed by the dynamic and context-specific nature of such risks. Many tests and validations have been conducted using diverse datasets acquired from social networking platforms. The findings of this study designate that the CGANSRI-ATSM framework exhibits the capacity to classify suicide risk in online settings at an initial phase with enhanced accuracy and recall associated to present organizations. Our results establish that the combination of CGANSRI-ATSM Fast text yields impressive recall, precision, and accuracy measures, with caps of 85.626%, 86.524%, and 94.637%, correspondingly. This research contributes to the field of suicide prevention by using DL models and feature engineering methods to analyze social media data. By leveraging these methods, we objective to develop suicide detection and prevention efforts in the context of the widespread use of social networking media.