Leveraging Rosat Classifiers for Career Succession Recommendation through Predicted Performance Ratings and XAI- SHAP for Features’ Influence Using HR Analytics

Abstract

Performance ratings of employees in an organization is a crucial factor for career succession. Predictive analytic algorithms, which uses historical human resource (HR) data, also known as HR Analytics, can be modelled to predict the performance ratings. Robust machine learning based classification algorithms are used to develop an ensemble of stacked classifier- RoSaT. This architecture is trained on real time data set and tested for predicting the performance ratings. Various metrics are used to assert the efficiency of the model. Finally based on the predictions, a recommendation list is generated, identifying employees with high performance ratings for potential career succession.