Employee Turnover Prediction by Machine Learning Techniques

Abstract

Employee turnover in Human Resource (HR) analytic is a term used to describe employees who leave the company due to termination, seek better job, or they are dealt with a bad working environment. Typically, a high turnover rate indicates that employees are dissatisfied with their current work environment. This leads to a high cost in terms of productivity,
time and money for the company as they were required to hire, rehire, and retrain the new employees to accustom themselves with their new work environment as well as the tasks assigned. In this paper, we propose a hybrid of machine learning algorithms and a Power BI model to design an Employee Turnover Prediction (ETP) application. Main factor influencing employee exit decisions and employee retention periods will be identified and the retention period for the employees or new applicants will be predicted. Employee dataset with the relevant features will be collected, processed, and analyzed. The analytics results (retention period) act as a benchmark for companies to determine whether they should hire applicants which also would possibly benefit to reduce the turnover rate of their company.