Employee Churn Prediction Using Machine Learning Techniques: A Systematic Review

One of the most important aspects of any planning is the part in which decisions are made. Employee attrition is that one part of an organisation, which demands a major chunk of decision making as it directly impacts the performance of a company irrespective of its size and geography. Hence, it is considered to be as one of the most well-known problems that require the focus of researchers so that right decisions are taken; thereby leading the administration to preserve high qualified employees. Interestingly, in recent times the domain of machine learning has grabbed the paradigm and has been witnessed as an efficient tool to predict such problems and warn which employee is more likely to quit. This procedure enables the human resources department to act quickly and perhaps offer a solution to the problem in order to stop attrition. Employee attrition is caused by a variety of factors. These factors are examined to show how they interact and highlight the predominate ones. This review's goal is to analyse different machine learning methods, including the decision tree (DT), XGBoost, AdaBoost, Gradient Boosting, Logistic Regression and KNN classifiers. Using the IBM Human Resource Analytic Employee Attrition and Performance dataset, these machine learning techniques are merged. The effectiveness of each of these supervised classification models for estimating employee turnover is analysed and developed using statistical methods through a systematic and detailed evaluation process. Furthermore, credible guidelines on the selection, application, and interpretation of these methods for the analysis of human resources datasets of various sizes and complexity are provided.