Intelligent Employee Retention System for Attrition Rate Analysis and Churn Prediction: An Ensemble Machine Learning and Multi- Criteria Decision-Making Approach

Abstract

The paper aims to examine the factors that influence employee attrition rate using the employee records dataset from kaggle.com. It also aims to establish the predictive power of Deep Learning for employee churn prediction over ensemble machine learning techniques like Random Forest and Gradient Boosting on real-time employee data from a mid-sized Fast-Moving Consumer Goods (FMCG) company. The results are further validated through a regression model and also by a multi-criteria Fuzzy Analytical Hierarchy Process (AHP) model which takes into account the relative variable importance and computes weights. The empirical results of the machine learning models indicate that Deep Neural Networks (91.2% accuracy) are a better predictor of churn than Random Forest and Gradient Boosting Algorithm (82.3% and 85.2% respectively). These findings provide useful insights for human resource (HR) managers in an organizational workplace context. The model when recalibrated by the human resource team of organizations helps in better incentivization and employee retention.