Suicidal ideation recognition based on sentence completion test via coding- and topic-enhanced model

Xuetao Tian, Jing Li, Xinyi Wang, Liang Xu, Fang Luo

Abstract: Highlights•This study introduces a novel perspective on suicidal ideation recognition through the Sentence Completion Test (SCT).•The study proposes a coding- and topic-enhanced model for suicidal ideation recognition based on a real-world dataset.•The study employs innovative methods for SCT analysis, e.g. diverse representations, focal loss, and contrastive learning.