| The current era is the rapid development of computer network technology and information explosion. Now more and more advanced equipment and abundant resources of the library. Library produces vast amounts of data in a business process during operation, it is only stored on the server, but does not reflect its value.This information has a lot of hidden data, valuable data worth mining.Through data mining techniques to tap the value of information. According exhumed the value of information can be achieved on readers a personalized service, which aims to change the way library services and improve the quality of library services.This paper analyzes and explains the subject of the personalized information service of the library, and discusses the necessity of the research on the implementation of the data mining technology in the library.In the paper, the research topics related to the theory of knowledge, which is mainly around the definition of data mining and development, mining process and tasks of the theoretical discussion. Including pretreatment of data backup, data mining, the establishment of a mining model, the results of expression and interpretation, in which to set up the data mining model is a crucial step in the data mining process, only the correct data mining model can dig out the results you need.This project mainly aims at the design purpose and principle of the personalized information service system. In the paper, according to the characteristics of the public library’s own characteristics, through the reader’s questionnaire survey to understand the real needs of the readers. This system combined with the data mining algorithm to analyze and solve the problem of library services in the transformation process will be faced with. Based on the theory of data mining, the personalized information service of the library is explored.This system is based on the library’s borrowing and recording data, there are a total of 10290 data. First the data pretreatment, find the information needed for mining data, get 5009 and data mining related data, the data reader refined classification, a total of 1780 and the reader’s related transaction data. Secondly, select the algorithm to meet the data characteristics and mining requirements, the paper mainly uses the K-means algorithm and Apriori algorithm, which is suitable for the implementation of data mining. In this paper, mainly focus on the analysis of the cluster analysis, the reader is divided into different groups, a group of targeted and targeted Association mining. Which the reader clustering process, and hope you can different types of readers’ reading behavior analysis derived from the final clustering results; the books borrowing data association rules mining, expect from the mining these data can be found on readers some interest and hobbies, and readers reading behavior preference. At last, according to the data mining, the relevant rules are obtained, and some suggestions are put forward according to the results of the mining. Through these suggestions, hope to give the correct decision support and guidance to the library manager.In this paper according to the understanding of the library work processes and the needs of readers of system design, through the technical support of SQL Server 2012 database, PHP, HTML established model of personalized information service system,n the system contains [my space], [personalized custom], [my borrow], [book recommendation] and other modules, to achieve the function of library personalized service.In this paper, the data mining technology to dig out the value of information from the hidden data. The value of information can help understand the needs of readers, for readers to provide convenient, improve the quality of service, realize the passive service of the traditional library change to active service; optimize the structure of the collection, shopping guide, improve library resources utilization, making library resources more reasonable and scientific. |
PDF Full Text Request