Multi-Label Classification of Employee Job Performance Prediction by DISC Personality

Abstract

The objective of this study was to automate job performance prediction based on DISC personality test. We transformed this problem to Multi-Label Classification (MLC) by using employee's job performances as labels. In this study, three widely used MLC techniques have been employed such as Binary Relevance (BR), Label Powerset (LP) and Classifier Chains (CC) for prediction of job performances. However, these traditional techniques didn't show promising results. Therefore, we proposed another approach by building stacking MLC with model selection. The proposed method has three steps: (1) building MLC model; (2) using process from the first step and applying with a stacking model and (3) utilizing feature selection technique to select the proper models for final prediction. Using the surveys from a big financial company in Thailand, we found that the last proposed approach shows better performance, compared to the traditional MLC.