Effective Classification Of Ibm Hr Analytics Employee Attrition Using Sampling Techniques

Today, in many software companies’ employees are quitting their jobs for a variety of reasons. When talented employees leave a good position, it becomes difficult for an organization to run a business. Therefore, organizations need to anticipate and analyze the reasons for termination of employees and develop appropriate plans and measures. IBM HR Analytics Employee Attrition and performance datasets are taken into account. In addition, there is an increasing need to fully understand the factors that influence attrition. Three sampling techniques were initially used in this paper: random oversampling, random undersampling, and SMOTE. In addition, the sampled dataset is sent to classification algorithms such as logistic regression, K-neighbor classifier, decision tree classifier, random forest classifier, and AdaBoost classifier for analysis of their performance metrics.