Identifying Suicidal Ideations

For identifying the underlying topics that express suicidal ideations, this study has employed the Latent Dirichlet Allocation (LDA) model. Semantic Network Analysis (SNA) is used to gain a deeper quantitative and qualitative insight into these texts. Besides, an exploratory investigation of different deep learning (DL) models has been performed to identify the posts speculating suicidal ideations. Furthermore, this study has integrated Explainable AI (XAI) techniques like Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) to enhance interpretability of the decisions taken by the DL models. Techniques like LDA and SNA offer a better understanding of the linguistic features of the suicidal posts, while the integration of the XAI techniques with the DL models elevates the transparency of their decisions.