Hybrid of K-means clustering and naive Bayes classifier for predicting performance of an employee

Abstract

Predicting the performance of an employee in the future is a requirement for companies to succeed. The employee is the organization's main component, the failure or organization’s success based on the performance of an employee, this has become an important interest in almost all types of companies for decision-makers and managers in the implementation of plans to find highly skilled employees correctly. Management thus becomes involved in the success of these employees. Particularly to guarantee that the right employee at the right time is assigned to the convenient job. The forecasting of analytics is a modern human resource trend. In the field of predictive analytics, data mining plays a useful role. To obtain a highly precise model, the proposed framework incorporates the K-Means clustering approach and the Naïve Bayes (NB) classification for better results in processing performance data of employees, implemented in WEKA, which enables personnel professionals and decision-makers to predict and optimize their employees' performance. The data were taken from the previous works, this was used as a test case to illustrate how the incorporates of K-Media and Naïve Bayes algorithms increases the exactness of employee performance predicting, compared with the K-Means and Naïve Bayes methods, the proposed framework increases the accuracy of predicting the performance of an employee.