Development of a job stress evaluation methodology using data mining and RSM

Data mining (DM) has emerged as one of the key features of many applications on information system. While a number of data computing and analyzing method to conduct a survey analysis for a job stress evaluation represent a significant advance in the type of analytical tools currently available, there are limitations to its capability such as dimensionality associated with many survey questions and quality of information. In order to address these limitations on the capabilities of data computing and analyzing methods, we propose an advanced survey analysis procedure incorporating DM into a statistical analysis, which can reduce dimensionality of the large data set, and which may provide detailed statistical relationships among the factors and interesting responses by utilizing response surface methodology (RSM). The primary objective of this paper is to show how DM techniques can be effectively applied into a survey analysis related to a job stress evaluation by applying a correlation-based feature selection (CBFS) method. This CBFS method can evaluate the worth of a subset including input factors by considering the individual predictive ability of each factor along with the degree of redundancy between pairs of input factors. Our numerical example clearly shows that the proposed procedure can efficiently find significant factors related to the interesting response by reducing dimensionality.