Study Of Fuzzy Clustering Algorithm And Its Validity

With the rapid development of the database technology and the wide application of DBMS, people have more and more data containing a great amount of valuable information. People want to deepen the analysis of the data, which helps people make better use of these data information. Now Database System can realize data input, search and statistics, etc, but it can't forecast the development trend of future data stored in the database. The short of measures to mine knowledge hiding behind the data results in the phenomenon which is that there is a large amount of data but poor knowledge. In the era of computer network, it has been a focus of attention that How we can obtain knowledge from large data effectively and rapidly. The abilities of data acquiring has been increasingly incompatible with the abilities of data analysis,so an automatic technology that can process data in a deeper level is needed. Data Mining is such a technology.As an important branch of data mining, clustering analysis, which can be an independent data-mining tool or 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. This paper has mainly studied partitioning-based fuzzy clustering algorithms and validity indices that are used to judge clustering validity. And also some research achievements are obtained.This dissertation is classified into five chapters:In Chapter 1, it is an exordium of the paper. Firstly, it introduces the research background and significances of this thesis. Secondly, it gives the thesis's research direction and achievements. Thirdly, it gives the introduction of clustering analysis and clustering algorithms. Finally, it gives the organization of the paper. In Chapter 2, it has mainly studied and discussed the research situation of fuzzy clustering algorithms, and gives a comprehensive survey of this area.