New dataset methodology aims to detect suicidal signals in Russian social media

Researchers have detailed a methodology for creating a large-scale dataset to identify presuicidal and anti-suicidal signals within Russian social media texts. This dataset, comprising over 50,000 annotated texts, aims to help identify individuals at risk by sifting through the vast amount of online content. The methodology covers instruction, annotation, verification, and correction processes, with the dataset, code, and materials made publicly available for further research and model development. AI

IMPACT This research could lead to improved AI models for mental health support by enabling better detection of at-risk individuals through social media analysis.

RANK_REASON The cluster describes a research paper detailing a methodology for dataset construction and includes preliminary experiments. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Russia
• ScienceCast

• paper
• safety
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →