Skip to content
Review Open access

Explainable AI (XAI): Making Artificial Intelligence Transparent

Oct 2026 · International Journal for Research in Applied Science and Engineering Technology · 0 citations

Abstract

Deep learning and ensemble models now drive decisions in healthcare, finance, and judicial systems, yet their internal reasoning remains largely opaque to the humans who rely on them. This opacity — the "black-box problem" — limits trust, complicates regulatory compliance, and makes it difficult to detect bias or error before harm occurs. Explainable Artificial Intelligence (XAI) addresses this gap by producing human-understandable justifications for model predictions. This paper reviews the major families of XAI techniques — intrinsically interpretable models, and post-hoc methods such as LIME, SHAP, Grad-CAM, counterfactual explanations, and attention-based interpretability — and proposes a hybrid explainability framework that combines global feature attribution (SHAP), local instance-level explanation (LIME), and rule-based surrogate extraction, evaluated through a novel Explanation Consistency Index (ECI). The framework is demonstrated on the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. The goal is to show that transparency and predictive performance need not be traded off against one another, and that hybrid explanation strategies can improve both trust and auditability of machine learning systems

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.