【作者】 徐海波；

【导师】 王建霞； 程普；

【作者基本信息】 河北科技大学 ， 计算机技术， 2010， 硕士

【摘要】 将数据挖掘技术应用于图书馆的个性化信息服务系统,挖掘隐藏在数据内部的潜在信息,能够更好的服务于读者的个性化信息需求。通过对数据库中的数据进行分析,从中发现隐藏的关系和模式,能够预测未来可能发生的行为,从而为管理决策提供有利的支持。数据挖掘技术使图书馆开展个性化信息服务成为可能。本文首先概述了课题研究背景和研究意义,其次介绍了图书馆个性化信息服务的相关理论以及数据挖掘技术的常用算法和常用工具,最后对个性化信息服务系统进行了整体设计,并对系统实现时的关键技术进行了详细介绍。本文研究工作的主要内容及关键技术如下：1)基于SSH框架采用数据层、挖掘层、业务层和用户层四层分层模型对个性化信息服务系统的整体架构进行设计,设计了系统数据采集、数据挖掘、个性化服务三大功能模块以及系统的主要功能具体包括我借阅的图书、我喜欢的图书、图书资源检索、最新图书推荐、好友推荐、虚拟参考咨询。2)对系统设计及实现时需要用到的关键技术：SSH框架整合技术、Weka数据挖掘软件、聚类分析算法K-MEANS进行了分析研究。3)基于数据挖掘构建读者兴趣模型和读者自建读者兴趣模型这种两种方式来构建读者个性化兴趣模型,并给出了具体的实现算法。4)利用SSH集成开发环境进行整合配置,以Weka作为数据挖掘工具,基于聚类分析算法K-MEANS实现了构建读者兴趣模型,了解读者个性需求,完成了图书推荐功能,从而实现了一个高校图书馆个性化信息服务系统。更多还原

【Abstract】 Through the analysis of data in the database. and found the hidden relationships and patterns. to predict possible future behavior. so as to provide favorable support for management decisions. Data mining technology allows library to develop personalized information services possible.This paper outlines the research background and significance. followed by the introduction of personalized information services. library-related theory and data mining tools commonly used algorithms and commonly used. the last of the personalized information service system for the overall design. and implementation When the key technology in detail.The main contents of this paper and key technologies are as follows:1) SSH frame is based on the data layer, mining layer, business layer and user layer hierarchical model of four personalized information service system on the overall architecture design. design a system of data collection. data mining. personalized service of three functional modules and specifically including the main function of the svstem to borrow my books. my favorite books. librarv resources search. the latest book recommendation. friend recommendation. virtual reference.2) The system design and implementation of the key technologies needed to use: SSH framework of the integration of technology. Weka data mining software. K-MEANS clustering algorithm was analyzed.3) Construction of readers interested in based on data mining models and readers interested in self-built model of the reader are two ways to build such a personal interest in the reader model. and gives a concrete algorithm.4) The use of integrated development environment to integrate SSH configured to Weka data mining as a tool. based on K-MEANS clustering algorithm to achieve the reader interested in building a model to understand the individual needs of the reader to complete the book recommendation feature. which implements a university library Museum of personalized information service system.更多还原