Bangla Suicidal Intention Dataset

Dataset Overview:The Bengali Suicidal Intention Dataset contains 13,288 Bengali (Bangla) text samples collected from publicly available social media posts, online comments, and blogs. Each text entry is labeled to indicate whether it reflects suicidal intention (1) or non-suicidal content (0). The dataset is designed for research in natural language processing (NLP), machine learning, and mental health studies, particularly for suicide risk detection in Bengali-speaking communities.

Columns:

Bangla – The Bengali text of the social media post or comment (String).

intention – Binary label indicating suicidal intention:

0 → Non-suicidal

1 → Suicidal

Data Shape: (13288, 2)

Number of rows: 13,288

Number of columns: 2

Temporal Coverage:

Start Date: 01/01/2025

End Date: 01/06/2025

Geospatial Coverage:Bangladesh, West Bengal (India), and other Bengali-speaking regions

Collection Methodology:The data was collected via web scraping and manually curated to ensure accuracy. Each sample was carefully labeled for suicidal intention by the author to maintain quality and consistency.

Intended Use:

Training and evaluating NLP models for suicide detection

Mental health research and monitoring

Ethical AI applications in Bengali-speaking communities

License:MIT License

Citation Suggestion:Jahangir Hussen. Bengali Suicidal Intention Dataset. Kaggle, 2025