【作者】 张通；

【导师】 王丹志；

【作者基本信息】 北京邮电大学 ， 计算机科学与技术， 2013， 硕士

【摘要】 随着近年来信息技术的飞速发展,我国的校园信息化进程也得到了极大的推动。学校教学和硬件资源不足等问题也随着办公自动化系统、信息管理系统、课程管理系统以及数字图书馆等现代化信息技术的发展而得到缓解。但随之而来的问题是高校图书馆的电子文献资源越来越多,高校内部信息系统呈现出信息孤岛的状态,缺乏连通性和联动性,通过整合高校内部信息,消除信息孤岛,提供综合性和个性化服务已经成为国内外数字化校园建设的发展趋势。本论文出自北京邮电大学青年科研创新计划专项中的《基于图书馆业务数据分析服务的学生个性化知识服务平台》课题。本论文的研究工作主要解决了学生个性化知识服务中的学生个性化推荐的需求。本论文以学生在图书馆的借阅、收藏、评价以及用户社会关系等数据为基础,深度挖掘学生个性化兴趣信息进而设计和实现了学生个性化信息推荐服务。在学生个性化信息推荐系统中,首先研究分析了现阶段比较成熟的推荐系统以及常用的推荐算法,并结合现有的比较成功的电子商务网站的推荐系统进行分析对比。然后设计了基于图书馆数据业务的学生个性化推荐系统的整体架构以及相关推荐引擎的架构设计,最后详细阐述了学生个性化推荐系统中的学生图书推荐模块、学生好友推荐模块、学生标签推荐模块的设计与实现。本论文主要贡献是将电子商务领域中已经相对比较完善的推荐系统思想与图书馆的数据业务相结合进而设计并实现了学生个性化推荐系统。本论文提出的推荐系统整体架构以及推荐引擎的思想符合软件工程模块化的思想,极大提升了系统的灵活性及可扩展性,另外在对系统中的推荐模块中结合不同的用户需求进而设计并实现的算法进行测试后可以看出,结合图书馆的数据业务特性最终实现的推荐结果基本满足了目前系统的需求。更多还原

【Abstract】 In recent years, science and information technology has a very rapid development, in this background, the construction of campus informatization has made great achievement in our country’s colleges and universities, and plays a pivotal role in the development and reform process of our colleges and universities. the emergence of major teaching system, such as student information management system, online courses system, student course management system, campus card system and digital library, etc, makes up for the weakness of the university hardware resources and teaching resources to a great extent. Attendant problem is the growing number of electronic literature resources of the university library, the university internal information system presents information island state, lacks of connectivity and linkage, through the integration of the internal information, eliminating the information island, to provide comprehensive and personalized service at home and abroad has become the development trend of the construction of digital campus.This paper uses the data of students’ borrowing, collection, evaluation and social relationship in the library as the foundation, excavate the students’ personalized interest information deeply, and then design and realize the students’ individualized information recommendation service. In students’ individualized information recommendation, recommend system mainly design and finished user items recommend module, user friend recommended module, user tag recommendation module items recommend module based on the label.The main function of the user items recommended module is to recommend users interested resources by Item-based collaborative filtering and user-based collaborative filtering, user’s friends recommended modules recommend interested friends for users according to the user’s attributes, interests, friends relations, ser tag recommendation module contains the user recommend tags and give users interested recommended resources based on the user’s label.更多还原