Employee Attrition Prediction Based on Gradient Boosting Approach

Abstract

In today's organizational landscape, predicting employee attrition has emerged as a critical concern. The departure of trained, technical, and pivotal staff members poses significant challenges, including financial setbacks incurred in their replacement. To address this, organizations harness current and historical employee data to discern prevalent attrition triggers. Employing established classification methodologies such as Decision Tree, Logistic Regression, Random Forest, Support vector machine, and Gradient boosting Algorithms are constructed using human resource data. Leveraging feature selection techniques, these models facilitate proactive measures to mitigate attrition risks. By accurately forecasting attrition, companies not only fortify their workforce stability but also enhance economic resilience through diminished human resource expenditures. This proactive approach not only aids in retaining valuable talent but also fosters sustainable growth by fostering an environment conducive to employee retention and organizational stability.