Assessment of Machine Learning Algorithms for Predicting Employee Attrition

In today’s world highly demanding and competitive work environment, precise insights into the causes of employee attrition which helps the organization to move on improving factors, so that they can retain their hardworking and excellent employees. Such insights will help the employees to absorb the nature of the organisation and their environment. By maintaining their vision constantly and improving the quality of service, the organisation is better informed to the employee, whom they intend to join. Based on attrition dataset, we have designed the system using several machine learning classifiers like Logistic regression, SVM, Random Forest, KNN and Decision Tree. Only accuracy rate metric of Decision tree is low compared to all metric in Random Forest. Therefore, we propose Improved Optimized Decision tree classifier to improve its accuracy. By comparing the metrics of these models, Improved Optimized Decision tree classifier performs better accuracy of 99.9%, precision of 99% and F1 score of 99%.