The Study And Application For Cluster Analysis Of Data Mining Method

With the development of information technology, Data Mining has been paid attention extensively. As we know, Data Mining has a large research scope, Cluster analysis is one of important research subject in it. Researching into the subject deeply has most important values not only on theoretic but also on applications. Clustering divides a set of physical or abstract object into several clusters that constitutes for the similar object, the object in the same of cluster are similar each other, but the object in the different cluster are dissimilitude each other. At present, clustering has been applied to pattern recognition, data analysis, image processing, and market research.Clustering is a very active research area in Data Mining. There are a great quantities of clustering algorithms in the documentation. To choose a algorithm is decided by the type of the data, the purpose and the application of the clustering. Clustering algorithms can branch out into the partitioning method, the hierarchical method, the density-based method, the grid-based method, and the model-based method etc. Among them, in medium and small scale application of data clustering, the partitioning method surpass to other methods in the easy comprehension, easy training, easy execution and good currency etc.On the base of above studies, in this paper, we probed a application of clustering technology in the teaching management of the high school. Through data analysis for"a table of hearing a lesson"of computer class. According to implement flow of clustering mining, we carried on all kinds of data preprocessing, we used the partitioning method's k-means algorithms to generate the first cluster, then used the replacement over and over to locate means in order to improving clustering result, in the end, made the square error minimum of each cluster minimum, completed clustering of teacher's"a table of hearing a lesson"using C++ program, then analyzed and verified the results of clustering, and concluded some results of having meaning practically to guide teaching management.