The Vague Sets Theory And Its Applictions In Clustering Analysis

Since the Fuzzy sets theory was presented by professor L.A.Zadel, it has been successfully applied in many fields. In real life, lots of uncertain information and data can not be accurately and overall expressed by Fuzzy sets theory, but Vague sets has more advantages over Fuzzy sets in expressing the "approve"、"oppose" and "uncertain" information. So it is of great importance to research Vague sets theory and its applications. The main contents in this paper are as follows:(1) In this paper, the relations of Vague sets and Fuzzy sets、the basic properties and methods of Vague sets are discussed. Based on the Vague sets theory and its voting model, the methods by which effectively transforming the Vague sets into Fuzzy sets are analyzed. At last, after fully considering the willingness of the neutral, a mew method is presented.(2) The clustering method, which is based on the equivalent relation, cause some noise of the original data, it is because this method increase the matrix between the operation. Though the method by directly using the similar relation is simple and can not cause the loss of data, it is a bit unreasonable in the choice of confidence level. To solve these problems, the concept of Vague optimal tree is presented and applied in the clustering.