ABSTRACT Human Resource Management has been transformed by Artificial Intelligence, by leveraging data-oriented analysis workforce analytics and use predictive analytics for Human Resource decision-making; and advances in machine-learning continue to enhance Human Resource capacity to be more accurate in analysing and predicting an employee's actions. The limitations of currently available research include the dependency on individual models, reduced reliability in making predictions, a lack of integration with decision support systems and as such, there is a gap in the way Human Resource decisions can be effectively made. The aim of this study is to develop the artificial intelligence design to predict when employees will leave, creating an efficient decision process within Human Resources. The workflow will use IBM's HR Analytics Employee Attrition data set for analytics, including preprocessing the data, extracting and selecting features, designing a Random Forest-based predictive framework, and then connecting this to a layer of decision-making optimization for use by HR in implementing their strategies. The experiment had successful results that showed an accuracy rate of 0.9352, precision rate of 0.9649, recall rate of 0.9016, F1-score of 0.9322, and a ROC-AUC of 0.9348, which is higher than the performance of conventional models like SVM, Decision Tree or KNN. The study shows that the proposed framework improves employee attrition prediction and supports better HR analytics for effective retention strategies.
Kun Jiang (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: