Rule Generation Based on Novel Two-stage Model

Purpose – The purpose of this paper is to develop a novel two-stage model for promoting the effect of rule generation based on rough set. In order to improve traditional rough set method, the novel two-stage model adopts new kernel intuitionistic fuzzy clustering (KIFCM) to promote performance of rough set theory. Moreover, the e-learning customer data set in Taiwan is also examined for demonstrate the effectiveness and practicality of model. Design/methodology/approach – In this paper, the authors present a new kernel intuitionistic fuzzy rough set model which combines novel KIFCM with rough set. The rule generation can divide to two stages for effective rule generation. In the first stage, KIFCM can utilize the advantages of kernel function and intuitionistic fuzzy sets to cluster raw data into similarity groups. In the second stage, the rough set theory is employed to generate rules with different groups. Finally, based on decision rules of rough set with different groups the results of system can be obtained and analyzed for users. Findings – The novel rule generation model adopts pre-process, which is KIFCM clustering technique, can effectively assist traditional rough set in promoting the performance. In analysis of e-learning data set, the empirical result indicates that proposed novel rule generation model can outperform traditional decision models.