Explainable AI in High-Risk Decision Systems: Improving Transparency in Healthcare and Banking

Abstract

Explainable Artificial Intelligence (XAI) aims to address these issues by providing clear and interpretable explanations for AI predictions. This paper presents a detailed study of XAI techniques and their applications in healthcare and banking. The proposed system integrates machine learning models with explanation methods such as LIME and SHAP to improve transparency and decision-making. The system enables users to understand model behavior, identify key influencing factors, and detect bias in predictions. The results demonstrate that XAI improves trust, fairness, and accountability in AI systems, making it suitable for deployment in critical real-world applications.

Ähnliche Arbeiten