Analyzing the Employee Attrition Rate: A Comparative Study of Various Machine Learning Approaches to Foresee Employee Attrition

Every organization's most valuable commodity has always been its workforce. Employees of a company have an enormous effect on its growth and innovation. Surprisingly, the Indian IT sector has recently experienced a historically high attrition rate. Any enterprise that experiences a high incidence of job turnover, risks losing competent workers as well as time and money. Employee Attrition also known as employee turnover occurs when any staff personnel leave the company either voluntarily or involuntarily. Several reasons, such as poor income and dissatisfaction at work, a lack of career opportunities, a toxic workplace culture, a lack of employee enthusiasm, etc., could lead an employee to quit his job. High staff turnover has historically been correlated with low unemployment. Employees are more likely to switch professions laterally as the number of available opportunities expands, seeking a better option, a higher salary, or a more appealing lifestyle. Therefore, employee attrition results in a massive loss for an organization. Predicting the attrition can help in strategic talent retention, or even a replacement can be held ready if required and in many more ways to retain the employee and reduce the attrition rate. The paper proposes the use of various EDAs to show the effect of different parameters on attrition. This work offers an in-depth comparison of numerous machine learning (ML) algorithms, including logistic regression, K Nearest Neighbours, Naive Bayes, Support Vector Machine, decision trees, and random forests. The results displayed will aid us in detecting the behavior employees who are likely to be attired in the future.