Employee Attrition Prediction

Abstract

Employees are the most valuable assets of an organization. To find, attract, develop and retain the right talent is a major part of management. Whenever a well-trained and well-adapted employee leaves the organization, it creates a vacuum. Therefore, the organization loses key skills, knowledge and business relationships. This study aims to identify attributes that contribute in employee attrition and numerical experiments are performed on these attributes using supervised learning methods like Support vector machine, Random Forest, Naive Bayes, Extreme gradient boosting. The performance of each of these supervised machine learning methods is analyzed through a robust and comprehensive evaluation process. This survey will help the human resource managers to identify the employees that are likely to leave the organization and predict the possible reasons for their decision, which will enable the HR managers to devise a retention plan or look for replacement.