Study On Some Applied Technologies Based On Rough Set Theory

Rough set theory is an excellent mathematical tool for dealing with vague, uncertain, imprecise information, it has been successfully applied in many fields. Nowadays, although much research has been taken on the side of rough set theory and application, it can not meet the need for solving the practical problems. In this paper, we consider the insufficiency about rough set for deal with uncertain problems and applications, classical rough set model is extended. Further more, fuzzy set theory, the theory of certainty, evidence theory are studied.Firstly, a new method is proposed for calculation the similarity about different attributers, namely, the fuzzy clustering decision based on the significance of attributers. By the proposed method, the different attributers are coupled (especially the quantitative attributes and the qualitative attributes) in the fuzzy information table or information table considering the significance of different attributers . In additions, the questions of clustering about quantitative attributes and qualitative attributes in information table are solved, and, the decision tables are set up accordance with the results of fuzzy cluster.Secondly, considering the shortcoming of classical rough set in rules acquisition, distributed Bayesian rough set model is proposed, and certainty rules are acquired using the proposed method, then the function of minimum risk is gained based on certainty rules. Next, Bayes decision of the minimum risk is used to classify a new data set, and an example is given. Thirdly, focus on the uncertain reasoning problem about expert system, the theory of certainty and its shortage are discussed, certainty factor is given by expert in theory of certainty which has subjectivity. So the model of reliability theory is improved on and the theory of certainty based on variable rough set is obtained. Finally, certainty factors are gained from information table, the objectivity about certainty factors is improved.Fourthly, the relation between evidence theory and rough set is discussed. The shortage about evidence theory for deal with uncertain problem, and evidence theory based on variable rough set is constructed. A case shows that this method is feasible and practical.Finally, a conclusion is given for this paper and simply describes the new ideas of this paper.