Explainable Artificial Intelligence for Workforce Decision-Making in Enterprise Systems

Artificial intelligence (AI) is now becoming a more popular option with regards to decision making related to workforce, such as talent acquisition, workforce employee retention prediction, performance assessment, and workforce planning. Nevertheless, the intensive usage of black-box models has triggered alarm among the stakeholders of the enterprise in the areas of transparency, trust, fairness, and compliance with regulations. This paper will discuss how Explainable Artificial Intelligence (XAI) can be used to overcome these problems in workforce prediction systems. A explainable machine learning system is built based on SHAP and local interpretability methods to offer significant results about the real-world workforce information, both on the global and features levels. According to the experimental findings, explainability is a key factor that enhances the level of expert trust in model outputs; in fact, the level of trust went up by more than 35 percent in contrast to non-explainable methods. Moreover, the suggested framework makes it possible to conduct successful human-AI interaction with the domain experts being able to examine and verify predictive results, with explainability being the key to responsible enterprise AI implementation.