A Novel Approach on Early Prediction of Employee Attrition in an Institution using Machine Learning Techniques

Abstract

Abstract: Institutions can only succeed if they have good employees. Retaining a good employee in an institution is a must to its growth. Sometimes, employees face issues in their institution because of overwork, no promotions, no rewards for good work, not seeing eye to eye with their manager, frequent business trips and extreme conditions which lead them to look for new jobs in the market. Employee attrition can be curbed if these causes are found sooner. To predict an employee’s resignation, Machine Learning Techniques are utilized. Attrition rates in an organization are predicted by factors such as work-life balance, opportunities, office atmosphere, pay, and other benefits. The Human Resources team will find the attrition rate data to be quite helpful in keeping exceptional employees. Random Forest, K-Nearest, Support Vector Machine and XG Boost are algorithms used to predict the attrition rate in an institution. The Human Resources Management (HRM) dataset is used by the models to detect various data aspects and efficiently estimate employee attrition.