A Hybrid Predictive Model for Early Detection of Employee Attrition in Organizations

Abstract

Employee attrition has resulted in the loss of tacit knowledge, high cost of hiring and morale of a team, which is a significant operational and financial challenge to the contemporary organizations. The traditional reactive plans to turnover are not sustainable in a dynamic business environment where active human capital management is significant. The paper will formulate and validate a hybrid predictive model that will predict employee attrition early on by utilizing the support of K-Nearest Neighbors (KNN) algorithm and Support Vector Machine (SVM) classifier. The proposed methodology involves the usage of a systematic data preprocessing pipeline, and feature engineering techniques in one of the synthetic human resources data sets, and stacked ensemble architecture.