Early Suicide Prediction

please cite this paper if you use this dataset:

Alom, M. S., Hoque Tomal, M. A., Taha, R., Parvez, S., Layek, M. A., Mohsin, M., & Talukder, M. A. (2025). An Explainable Triple-Layered Ensemble Model for Early Prediction of Suicide Risk Using Machine Learning. Engineering Reports, 8(1), e70574. https://doi.org/10.1002/eng2.70574

paper link: https://onlinelibrary.wiley.com/doi/10.1002/eng2.70574

📊 Dataset Description: Final Student Psychological Dataset
This dataset is a curated collection focused on exploring the mental health status of students, with a particular emphasis on stress, depression, anxiety, and suicidal tendencies. It can be used for mental health risk prediction, early intervention modeling, and behavioral pattern analysis using machine learning techniques.

🗂️ Dataset Summary
Rows: 1,099

Columns: 12

Missing Values: None

File Format: CSV

📌 Column Descriptions
Column Name Description
Age Age of the student (Integer)
Gender Gender identity (e.g., Male, Female)
Stress Level Level of stress reported (e.g., Low, Moderate, High)
Academic Performance Self-reported academic performance (e.g., Good, Poor, Average)
Health Condition General physical health (e.g., Normal, Fair, Abnormal)
Relationship Condition Current relationship status (e.g., Single, In a relationship, Breakup)
Family Problem Any existing family issues (e.g., None, Financial, Parental conflict)
Depression Level Level of depression (e.g., Sometimes, Often, Always)
Anxiety Level Level of anxiety (e.g., Sometimes, Often, Always)
Mental Support Type of support system (e.g., Family, Friends, Loneliness)
Self Harm Story Past self-harm experience (Yes/No)
Suicide Attempt History or thought of suicide (e.g., Never Thought, Thought, Attempted)

💡 Potential Use Cases
Suicide Risk Prediction using classification algorithms.
Mental Health Monitoring across student populations.
Behavioral Health Trend Analysis in educational institutions.
Stress and Depression Correlation Studies.