Employee Churn Prediction Model using C4.5 Classification Algorithm

Churn phenomenon commonly happen in customer problem and become jeopardy issues that any industries can suffer. Churn problem also can appear in organization, it is called employee churn. Employee churn is relatable with customer churn yet slightly distinct. Churn create a numerous adversely effects in the organization such as loss of employee can lead to unfairly distribution of workload, customer dissatisfaction, also company costs money and time for finding a replacement. Hence, it is important to know who, where, and why the employee is churning. Classification and prediction in data mining is implemented to predict the employee churn. Therefore, this research aims to present a case study that we present a study of C4.5 classifier algorithm for employee churn prediction model. In the prediction proposed model, the splitting of training and testing data distinguish into 2 different types of ratios. For dataset 1 the training dataset is 70 % and testing dataset is 30 %, while for dataset 2 training dataset is 80 % and testing dataset is 20 %. The classifier accuracy for dataset 1 and dataset 2 gains 94.8 % and 95 % respectively. Based on the accuracy level, C4.5 classifier is the proper method to predict employee churn.