A Proposed Model for Predicting Employee Turnover of Information Technology Specialists Using Data Mining Techniques


  • Ahmed Ghazi Helwan University, Faculty of Commerce & Business Administration, Department of Business information systems, Cairo, Egypt
  • Samir Elsayed Helwan University, Faculty of Commerce & Business Administration, Accounting Department, Cairo, Egypt
  • Ayman Khedr University of Jeddah, College of Computing and Information Technology at Khulais, Department of Information Systems, Jeddah, Saudi Arabia




Data Mining Techniques, Prediction, Classification, Employees satisfactions and turnover


This article proposes a data mining framework to predict the significant explanations of employee turn-over problems. Using Support vector machine, decision tree, deep learning, random forest, and other classification algorithms, the authors propose features prediction framework to determine the influencing factors of employee turn-over problem. The proposed framework categorizes a set of historical behavior such as years at company, over time, performance rating, years since last promotion, and total working years. The proposed framework also classifies demographics features such as Age, Monthly Income, and Distance from Home, Marital Status, Education, and Gender. It also uses attitudinal employee characteristics to determine the reasons for employee turnover in the information technology sector. It has been found that the monthly rate, overtime, and employee age are the most significant factors which cause employee turnover.






Review Papers