Employee Classification for Personalized Professional Training Using Machine Learning Techniques and SMOTE
Theeramet Kaewwiset, Punnarumol Temdee, Thongchai Yooyativong · 2021
Training and development are essential parts of professional development that are necessary for employees to improve their capacity. Generally, professional development program is organized based on personal information, such as background, personal goal, and work experience together with business objectives and job criterion. To promote personalized training in professional development process, the proper classification of individual employee is necessary. This paper thus proposes the classification method for employee classification to promote personalized training in organizations. The machine learning based methods, Decision Tree, Random Forest, and Support Vector Machine, are studied. Synthetic Minority Oversampling Technique (SMOTE) method is used to deal with imbalance data. The open data from Kaggle is used in this paper. For method validation, the data for training and testing are formed into three groups including 80:20, 70:30, and 60:40 respectively. The classification results show that the SMOTE can improve classification performance for all classifiers. Additionally, Random Forest performs the best classification with 96.32% accuracy.