BIAS DETECTION AND MITIGATION USING EXPLAINABLE AI IN INTELLIGENT RECRUITMENT AUTOMATION

The increasing adoption of Artificial Intelligence (AI) in recruitment automation has transformed hiring processes by enabling efficient candidate screening, ranking, and shortlisting. However, AI-driven recruitment systems may unintentionally inherit or amplify societal biases present in historical hiring data, leading to unfair decision-making and discrimination. This paper proposes a Bias Detection and Mitigation Framework that integrates Explainable AI (XAI) techniques into intelligent recruitment automation systems. The proposed model employs machine learning algorithms for candidate evaluation while incorporating fairness-aware metrics to detect demographic bias across protected attributes such as gender and ethnicity. Explainability tools such as SHAP (SHapley Additive exPlanations) are used to interpret model decisions and identify biased feature contributions. A bias mitigation module applies pre-processing and in-processing fairness correction strategies to improve equitable outcomes. Experimental evaluation demonstrates improved fairness metrics, reduced disparate impact, and maintained predictive performance compared to conventional AI-based recruitment systems. The proposed framework enhances transparency, accountability, and ethical compliance in automated hiring processes