基于文献引证关系的数字图书馆协同推荐系统  

Digital library collaborative filtering recommender system based on citation relations

作　　者：严海兵[1] 刘昊[1] 

机构地区：[1]苏州科技学院图书馆,江苏苏州215009

出　　处：《苏州科技学院学报（自然科学版）》2014年第3期75-80,共6页Journal of Suzhou University of Science and Technology （Natural Science Edition）

摘　　要：协同过滤算法应用于数字图书馆推荐系统有诸多优点,然而在系统启动初期存在最初评价问题和数据稀疏问题。针对这些情况,提出引入文献引证关系于数字图书馆协同过滤算法。文献引证关系能够较好的反映文献作者对参考文献、引证文献的兴趣评价。基于文献引证关系的数字图书馆协同过滤算法引入作者的文献评价信息,解决了协同过滤算法固有的问题。实验结果表明,该算法能有效地提高系统的推荐质量。Collaborative filtering algorithm in digital library recommender system has many advantages, but in the initial system startup, there were two major problems, including the initial evaluation and data sparsity. In view of this, we propose to introduce the literature citation relations to the digital library collaborative filtering algorithm. The relations can represent the literature evaluation of authors interested in reference of the literature and citation of the literature. Digital library collaborative filtering recommender system based on citation relations solves the inherent problems of collaborative filtering algorithm by introducing evaluations of the authors. The experimental results show that the algorithm can effectively improve the quality of the recommender system.

关 键 词：推荐系统 协同过滤 数字图书馆 推荐算法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]