Explainable AI System for Interpreting Machine Learning Predictions using LIME and SHAP

This paper presents a proof-of-concept system for Explainable Artificial Intelligence (XAI) using Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive explanations (SHAP). The system is designed to interpret predictions made by black-box machine learning models in two real-world scenarios: house price prediction (regression) and loan approval prediction (classification). A baseline machine learning model is trained for each task, and both LIME and SHAP are applied to generate local and global explanations. The system demonstrates how feature contributions influence predictions and converts them into human-readable explanations. The results show that SHAP provides consistent global insights, while LIME offers intuitive local interpretations. The proposed approach enhances transparency, trust, and interpretability in AI systems.