PREDICTION OF EMPLOYEE ATTRITION USING BAYES RISK POST-PRUNING IN DECISION TREE

Abstract

Employee attrition is the departure of employees from the organization for any reason (voluntary or involuntary), including resignation, termination, death, or retirement. Attrition is widely understood to be one of the major problems affecting organizations today. Losing employees has many direct and indirect impacts across a company. It occurs employee attrition when an employee leaves and is not replaced at all or for a significant amount of time, resulting in a reduction of the workforce. In this system, Decision Tree (ID3) classifier is used to analyze the causes of employee attrition. And then Bayes Risk Post-Pruning (PBMR) technique is applied to reduce the condition of overfitting on decision tree. The proposed system performance is evaluated various evaluation standards such as precision, sensitivity and F1 score values based on IBM Human Resource Analytic Employee Attrition and Performance dataset from Kaggle site. The proposed system compares the accuracy between before post-pruning and after Bayes Risk post pruning was applied. The proposed approach findings help organizations overcome employee attrition by improving the factors that cause attrition. This system is implemented by using python programming language with Google Collab Drive.