CONFERENCE / ICCAIS-2026

Mental health issues such as depression, anxiety, and suicidal tendencies rarely come to the attention of experts until they have reached an advanced stage. This is mainly because there has been minimal monitoring and is still an issue that attracts deep social stigma. On the other hand, there are many in- dividuals who are not afraid to reveal their feelings, thoughts, and states of mind through various social media platforms. This calls for the need to develop an Artificial Intelligence-based system that can help identify individuals who are struggling mentally. This paper seeks to develop an Artificial Intelligence system that uses various Natural Language Processing techniques and deep learning algorithms to determine and predict individuals who are struggling mentally. The developed model is expected to incorporate various features from recent Artificial Intelligence research undertaken to predict crisis situations and conditions using linguistic and behavioral cues as well as temporal cues. The hybrid deep learning model proposed is expected to work by incorporating Bidirectional Encoder Representations from Transformers and Bidirectional Long Short-Term Memory to enable the system to achieve high accuracy in its predictions.