Research on Mining Rules from Multi-criterion Group Decision Making Based on Genetic Algorithms

Abstract

Multiple Criterion Group Decision Making (MCGDM) which is based on the procedure has the virtue of drawing on the wisdom of masses with the defect of time and resource wasting. Then the experience of the historical MCGDM processes as the collective knowledge for future tasks is possible to be made use of to overcome the shortcoming of MCGDM without losing its advantage. But the existing techniques seldom handle the linguistic data in MCGDM as the knowledge availably. In this article, we propose a method of mining the briefest rules as the group experience from the decision table built from the historical MCGDM process. This method is based on genetic algorithms, which is designed by us. And the whole model is integrated in our prototype of knowledge oriented group decision support system and shows good impact on the instance.