AI model predicts male domestic violence in Bangladesh using explainable learning

Researchers have developed an explainable ensemble learning model to predict male domestic violence (MDV) in Bangladesh. The study addresses challenges of imbalanced data and limited availability by collecting data from nine major cities. Their proposed stacking ensemble model, using ANN and CatBoost as base classifiers with Logistic Regression as the meta-model, achieved 95% accuracy and a 99.29% AUC. Explainable AI techniques like SHAP and LIME were employed to provide transparency into the model's decision-making processes, challenging the notion that domestic abuse primarily affects women and highlighting the need for tailored support for male victims. AI

IMPACT Develops new methods for applying explainable AI to sensitive social issues, potentially improving understanding and intervention strategies.

RANK_REASON Academic paper detailing a novel application of machine learning and explainable AI techniques to a social issue. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• artificial neural network
• Bangladesh
• Catboost
• deep learning
• logistic regression model
• machine learning
• Male Domestic Violence
• Md Abrar Jahin
• Shap

• paper
• safety
• other

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