Predicting Employee Attrition: A Machine Learning Approach with Interactive Dashboards

Abstract

Organisations increasingly use data-driven insights to guide their decision-making in the modern digital world. This study investigates employee attrition, a problem that affects businesses everywhere and can result in everything from higher hiring expenses to service interruptions. This study aimed to forecast staff attrition rates and give visual dashboards to help HR departments with strategic planning by utilising the "IBM HR Analytics Employee Attrition & Performance" dataset from Kaggle. The dataset was analysed using machine learning techniques, notably decision trees and random forests. With Tableau, a dynamic dashboard emphasising user interaction and interactivity was created. This dashboard displayed the outcomes of the prediction models and gave users access to information about the many aspects that affect employee attrition. Protecting individual rights and building faith in the analytical results were ensured by addressing legal, social, and ethical issues, particularly in data processing. According to the results, the random forest algorithm fared better in predicting staff attrition than decision trees. The interactive dashboard's capabilities, which include filters for age group, income comparison, job role, and department, improved the user experience and gave users a more detailed understanding of attrition trends. This study highlights the potential of data analytics in HR management by providing tools and insights that can significantly impact organisational strategy and decision-making.