Predicting Stress in Bangladeshi University Students: A LIME-Interpretable Machine Learning Approach
Description
This dataset was collected as part of the study: Predicting Stress in Bangladeshi University Students: A LIME-Interpretable Machine Learning Approach (https://www.researchgate.net/publication/383053109_Predicting_Stress_in_Bangladeshi_University_Students_A_LIME-Interpretable_Machine_Learning_Approach) Description The dataset contains responses from 571 students from various public, private, and national universities in Bangladesh, collected between September 2022 and January 2023 via an online survey (Google Form). It is designed to predict and analyze stress levels among university students using: DASS-21 (Bangla Version) – for stress level assessment Ten-Item Personality Inventory (TIPI) – for personality traits Additional demographic and lifestyle factors Structure Demographics – Academic year, age, CGPA, university type, etc. Stress Factors – Living arrangements, academic satisfaction, financial difficulties, exposure to violence, bullying, relationship issues, social media use, etc. DASS-21 Items – 7 questions (Q1A–Q7A) TIPI Items – 10 questions (TIPI1–TIPI10) Target Variable – Stress level: Normal (0–18) Moderate (19–25) Severe (26+) Potential Uses Stress level classification Personality–stress correlation analysis Feature importance analysis with XAI (LIME, SHAP) Academic performance impact analysis Citation If you use this dataset, please cite: Hosen, M.H., Islam, M.T., Ashraf, K., & Haque, P. (2024). Predicting Stress in Bangladeshi University Students: A LIME-Interpretable Machine Learning Approach. ResearchGate. https://doi.org/10.13140/RG.2.2.32712.62727 Ask ChatGPT
Files
Institutions
- East Delta University