高校图书馆图书推荐系统中的稀疏性问题实证探析  

Empirical Study on Sparsity of University Libraries

作　　者：张闪闪[1] 黄鹏[2] 

机构地区：[1]中国科学院文献情报中心,北京100190 [2]北京师范大学政府管理学院,北京100875

出　　处：《大学图书馆学报》2014年第6期47-53,共7页Journal of Academic Libraries

摘　　要：分析高校图书馆现有图书推荐系统存在的自动化和稀疏性问题,并结合图书馆的特色,提出通过建立自动化评分规则,结合中图法类目体系进行图书归类,形成新的书目数据库的方法来实现图书的推荐计算。最后,通过向300名学生推荐图书,实证分析了自动化评分规则和中图法类目在高校图书馆图书推荐中的应用效果。This paper analyzed the problem of automation and sparsity existing in the personalized recommendation service of university library. It proposed new method to achieve recommended calculation of books through an automated scoring system, which combined the characteristic of library and categorized the books by fourth edition Chinese Library Classification to form a new bibliographic database. Finally, the authors did an experiment of which the results show that the new algorithm achieves feasibility and improved the accuracy of book recommendation significantly.

关 键 词：高校图书馆 稀疏性 自动化评分 可行性验证 书目数据库 

分 类 号：G258.6[文化科学—图书馆学]