Application Of Data Mining In University Library Management System

The rapid development of computer technology, Internet technology and information technology, to people's production and life brought great changes, but also brought profound changes to the field of library. For a long time,the library has been the paper books as the most important resources. However, with the development of information technology, the paper books to cover more comprehensive knowledge, we need more library space, also to the preservation of books, borrowing and returning management cause the enormous pressure. In this case, universities, research institutions and other units have begun to build electronic libraries.Electronic library so far there has indisputable fact confirmed the will become the major mode in the field of library. But the electronic library also gives developers a series of new challenges, the most important question is how to quickly and accurately retrieve the user's needs in the vast amount of information in the book. Mass storage is an important advantage of the electronic library, but the greater the amount of storage also increased the difficulty of the search process. In this situation,data mining has become the first choice to improve the retrieval performance and improve the quality of electronic library services.In this paper,based on data mining technology of the electronic library for the study,the following study:First, the basic theory of data mining are described, the typical data mining algorithms are outlined, including association rules algorithm and regression analysis algorithm,Bayesian algorithm, decision tree algorithm, neural network algorithm, genetic algorithm.Secondly,the typical methods of cluster analysis are analyzed. Based on the K-Means means clustering method, the data of the users in the electronic library and the data of the library data is tested, and the validity of the method is verified.Thirdly,according to the shortcomings of Apriori algorithm,an improved Apriori algorithm is proposed. Experimental results show that the improved Apriori algorithm has more efficient data mining performance.Finally, the electronic library management information system were system structure design and system function design; the database ER model design, data table design of the electronic library management information system is; the realization of the system is given and the key function for the show.