Monalisa Jena1, Noman Khan2,*, Mi Young Lee3,*, Seungmin Rho3
CMES-Computer Modeling in Engineering & Sciences, Vol.146, No.1, 2026, DOI:10.32604/cmes.2025.075119
- 29 January 2026
Abstract Mental-health risk detection seeks early signs of distress from social media posts and clinical transcripts to enable timely intervention before crises. When such risks go undetected, consequences can escalate to self-harm, long-term disability, reduced productivity, and significant societal and economic burden. Despite recent advances, detecting risk from online text remains challenging due to heterogeneous language, evolving semantics, and the sequential emergence of new datasets. Effective solutions must encode clinically meaningful cues, reason about causal relations, and adapt to new domains without forgetting prior knowledge. To address these challenges, this paper presents a Continual Neuro-Symbolic Graph… More >
Login
Register
