Artificial Intelligence Techniques to Predict Employee Satisfaction: Strategic Implications for Human Resource Management

A Predictive HR Analytics Approach Using XGBoost and Explainable Artificial Intelligence

Abstract

Employee satisfaction is a critical factor influencing employee well-being, organizational performance, productivity, and workforce retention. Traditional approaches for assessing employee satisfaction primarily rely on retrospective surveys and descriptive analyses, limiting organizations' ability to proactively identify dissatisfaction and implement timely interventions. This study aims to develop an explainable machine learning framework for predicting employee satisfaction using structured human resource data and to identify the key factors influencing employee satisfaction.

A publicly available HR analytics dataset containing 15,787 employee records was utilized. Four machine learning classification algorithms, namely Logistic Regression, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost), were developed and evaluated. To enhance model transparency and support managerial decision-making, Explainable Artificial Intelligence (XAI) techniques based on SHapley Additive exPlanations (SHAP) were employed to identify the most influential predictors of employee satisfaction.

The experimental results indicate that XGBoost achieved the highest predictive performance, with an accuracy of 98.3% and an F1-score of 0.98. SHAP analysis revealed that work-life balance, salary, promotion history, and years at the company were the most influential factors affecting employee satisfaction. The findings demonstrate that employee satisfaction is influenced by complex interactions among organizational and career-related factors that can be effectively captured through advanced machine learning techniques.

This study contributes to the growing field of HR analytics by integrating predictive modeling with explainable artificial intelligence to support transparent and data-driven human resource management. The proposed framework can assist organizations in proactively identifying potential dissatisfaction, designing targeted interventions, and enhancing strategic workforce management and employee well-being.

Keywords: Employee Satisfaction, Human Resource Analytics, Machine Learning, Explainable Artificial Intelligence, XGBoost, Human Resource Management.