HR ANALYTICS FOR EMPLOYEE ATTRITION PREDICTION USING MACHINE LEARNING

Abstract

The practice of gathering and analyzing Human Resource data to enhance an organization's workforce performance is known as HR analytics. Employee data is used in HR analytics to help the firm make better decisions for future expansion. The term "attrition" refers to when employees leave the organization either voluntarily or involuntarily. This article uses machine learning and data analysis methods to examine employee attrition. The procedures include addressing missing numbers, visualizing the data, and using classification algorithms. The likelihood that an employee will leave the organization is predicted using different models, including Random Forest, Decision Tree, SVC, and Logistic Regression. Performance metrics such as the confusion matrix, accuracy score, precision score, and ROC Curve were utilized to evaluate how well the prediction was made. Our research led us to the conclusion that the Random Forest Classifier had the highest accuracy, with a 98% accuracy rate. This study's main objective is to retain employees by understanding the factors responsible for the employee to leave the company.