Predicting Employee Attrition using an Ensemble Method

Abstract

Employee attrition poses significant challenges for organizations, impacting productivity, performance, and financial stability. Existing research on predicting attrition suffers from limitations such as accuracy constraints, insensitivity, and a lack of finding the important features contributing to employee attrition. This study aims to address these gaps by developing an ensemble model for predicting attrition and enhancing employee retention rates. The objectives include improving prediction accuracy, identifying informative factors, dealing with imbalanced data, and incorporating hyperparameter tuning. The proposed ensemble model, augmented with hyperparameter tuning, achieved impressive performance metrics, including an accuracy rate of 92.75%, precision score of 98%, recall rate of 88.83%, specificity of 97.68%, and F-beta score of 93.25%. These results indicate the model's effectiveness in identifying employees at risk of attrition and its potential for aiding organizations in retention efforts.