HR RECOMMENDER SYSTEM FOR THE IMPROVEMENT OF EMPLOYEE ATTRITION USING DECISION TREE AND WHALE OPTIMIZATION ALGORITHM

Abstract

This paper presents the design and development of an advanced HR recommender system aimed at mitigating employee attrition; a critical challenge faced by organizations. Employee turnover incurs significant financial costs and negatively impacts productivity and morale. The proposed system combines Decision Tree Analysis with the Whale Optimization Algorithm (WOA) to enhance predictive accuracy in identifying employees at risk of leaving. Decision Tree Analysis serves as a classification tool to analyse historical employee data, pinpointing key factors contributing to attrition, such as job satisfaction, work-life balance, and career growth. Traditional decision tree models often struggle with overfitting and suboptimal performance. To address these issues, WOA, inspired by the hunting strategies of humpback whales, fine-tunes the decision tree model’s hyperparameters, improving prediction accuracy and generalization. The system utilizes data from employee surveys, performance evaluations, and organizational metrics, providing a comprehensive view of influencing factors. By leveraging this hybrid model, the HR recommender system predicts potential employee departures and suggests targeted retention strategies, enabling proactive measures such as training and compensation adjustments. The study demonstrates that integrating Decision Tree Analysis with WOA significantly enhances predictive performance, offering a valuable tool for effective employee retention management.