The Research Of Clustering Algorithm For Interval-Valued Intuitionistic Fuzzy Sets

As an important branch of data mining,clustering analysis,which can be an independent data-mining tool of preprocessing procedures of other data-mining algorithms,is attracting wide attention.Clustering is an unsupervised classification,and it is an important method with which people know the society and nature.In the real world,the boundaries between objects are usually fuzzy when we classify them.Therefore,the fuzzy clustering technology has been brought forward.The technology of data mining is a cross subject developing rapidly domestically and overseas,It is a process that to deal with and extract unknown information and knowledge from large databases by using the means of database,statistics,artificial intelligence,and maching learing.It has been applied extensively in the fields of agriculture,finance,insurance,national defence and so on.In recent years,people' researches in the intuitionistic fuzzy sets and interval-valued intuitionistic fuzzy sets(IVIFS) have been a hot topic.Fact show that intuitionistic fuzzy sets and interval-valued intuitionistic fuzzy sets can describe and portray the fuzzy character of objective word more delicately.However,so far there have been little investigation of clustering techniques of interval-valued intuitionistic fuzzy sets.The clustering analysis for interval-valued intuitionistic fuzzy sets is a new research field and the research will have great pratical significance. The paper has studied characteristic properties of interval-valued intuitionistic fuzzy sets, proposed the clustering algorithm for interval-valued intuitionistic fuzzy sets based on these properties.And also some research achievements are obtained.Basing on the interval-valued intuitionistic fuzzy basic theories,The interval-valued intuitionistic fuzzy relations, interval-valued intuitionistic fuzzy similarity relations and interval-valued intuitionistic fuzzy equivalent relations are introduced, with the emphasis on investigating their fundamental rules of operations, characteristic properties and related theorems,and an approach to transforming the interval-valued intuitionistic fuzzy similarity matrices into the interval-valued intuitionistic fuzzy equivalence matrices is given.λ-cutting matrix of the interval-valued intuitionistic fuzzy equivalent matrix is given,and then an approach to clustering interval-valued intuitionistic fuzzy sets is proposed.An example is given to illustrate and analyze the approach. A developed approach to describes the degree of dissimilarity between interval-valued intuitionistic fuzzy sets is proposed using four definite physical significances. The advantage of the approach is that it can express the degree of dissimilarity between interval-valued intuitionistic fuzzy sets more accurately,and retain more information.Then,the similarity between interval-valued intuitionistic fuzzy sets is introdeced.Basing on the upper theories,the paper finally study the clustering algorithm for Multi-properties interval-valued intuitionistic fuzzy sets.