Early Prediction of Employee Turnover Using Machine Learning Algorithms

Employee turnover is a serious challenge for organizations and companies. Thus, the prediction of employee vital issue in all The present work proposes models predicting intentions workers during recruitment process. proposed are based on k-nearest neighbors (KNN) random forests (RF) machine learning algorithms. use dataset created by IBM. used includes most essential features, which considered process may lead to turnover. These features salary, age, distance from home, marital status, gender. KNN-based model exhibited better performance terms accuracy, precision, F-score, specificity (SP), false-positive rate (FPR) comparison RF-based model. predict average probability percentage workers. Therefore, can be aid human resource managers make precautionary decisions; whether candidate likely stay or leave job, depending given relevant information about employee.