The Study Of Clustering Algorithm Based On Books Management System

By the end of last century, Data Mining (DM) has become a kind of significant intelligent data analysis technology, with which the helpful knowledge of pattern will be drawn out or discovered automatically from vast various types of data which saved in databases or data warehouses. In this process, Cluster analysis is the important topic in the research field of data mining.In the existent system of libraries, plenty of data are generated every day, every month and every year. Facing these data, the key question is how to classify them. Clustering analyzing is a method of classifying data reasonably, its aim is to put the similar things together and the similarity is big in the same clustering. Therefore this article applies the k-means cluster algorithm in"the books management system", facing these mass data, how studies to classify reasonably them.First, this article introduced briefly data mining's elementary theory, then described the cluster meaning as well as in view of the different data type difference computation, then listed in the data mining the extant several kinds to have the representative cluster algorithm thought as well as their good and bad points, finally described the K-means algorithm with emphasis, as well as this algorithm basic philosophy and principle.Then, introduced in detail the books management system's analysis design with realizes, in the books manage, the major function include: The books increase, the inquiry, the browsing, the deletion and the revision manage of the books user's management, like this might facilitate the librarian to carry on the management to the huge information, this article uses had the platform to be small, safe, facilitated and so on handling characteristics JAVA languages to take the development tool. And for the information data, should have the characteristics of prompt memory, and prompt deletion, this need to have a security, the highly effective database to realize, therefore in this design, used the ACCESS 2003 databases to take the memory database, realized the function of books management.Finally it applies the typical k-meanscluster algorithm in"the books management system", carries on the cluster analysis to the books. This system takes the record of history borrows, the reader database as the essential originate data of"the books management system", with the k-means cluster algorithm analysis reader's courtyard, the sexual identity and the grade characteristic r separately, then according to ratio of the books utilization divides the books service condition into three kinds of the high, middle and low, obtains the cluster result finally. The cluster result indicated that the library Purchase unit should act according to the existing books to purchase suitably some societies books, the love story books, thrilling books, the deductive books, English reading books, the English basic vocabulary books, English conversation books, English test books, the Britain short story, the prose class books, the computer program language books and so on network class books meet the different student reader's need. The cluster result can instruct the library Purchase unit supplement books prompt to a certain extent, satisfies the books demand of the reader.