[Purpose/significance] This paper proposes a recommendation algorithm based on user interest metrics and content analysis for the current issues of low personalization and poor recommendation in knowledge discovery services. [Method/process] Through characteristic word distribution, LDA topic distribution and citation association, this paper constructs the academic resource model. Through the measurement of user behavior (browsing time, downloading, forwarding, collecting, etc.), the user's interest in browsing academic resources can be calculated, and the user interest model is constructed. Matching the user interest model with the academic resource model and calculating its similarity, the user's interest value for each academic resource can be obtained. Finally, the TOP-N academic resources with the highest interest value can be recommended to the user. [Result/conclusion] The paper tests the effectiveness of the algorithm and the accuracy of the recommendation through experiments. From the experimental results, we can show that the recommendation algorithm can predict the user's interest better and the recommendation effect is significant, simultaneously providing ideas for precise recommendation of discovery services.
Ding Mengxiao
,
Bi Qiang
,
Xu Pengcheng
,
Li Jie
,
Mu Dongmei
. Research on Precise Recommendation of Knowledge Discovery Services Based on Users Interests[J]. Library and Information Service, 2019
, 63(3)
: 21
-29
.
DOI: 10.13266/j.issn.0252-3116.2019.03.003
[1] 毕强,刘健.基于领域本体的数字文献资源聚合及服务推荐方法研究[J].情报学报,2017,36(5):452-460.
[2] WALTERS W H. Google Scholar coverage of a multidisciplinary field[J]. Information processing & management, 2007, 43(4):1121-1132.
[3] YANG S Q, WAGNER K. Evaluating and comparing discovery tools:how close are we towards next generation catalog?[J]. Library hi tech, 2010, 28(4):690-709.
[4] MICHAEL G. The evaluation of discovery services at Lynchburg College:2009-2010[J]. College & undergraduate libraries, 2012, 19(2-4):387-397.
[5] 秦红.普适计算环境中的数字资源感知服务框架探讨[J].图书情报工作,2014,58(5):13-16,21.
[6] 张钧.基于用户画像的图书馆知识发现服务研究[J].图书与情报,2017(6):60-63.
[7] GUAN P, WANG Y F. Personalized scientific literature recommendation based on user's research interest[C]//International conference on natural computation, Fuzzy systems and knowledge discovery. Changsha:IEEE, 2016:1273-1277.
[8] RICCI F, ROKACH L, SHAPIRA B, et al. Recommender systems handbook[M]. New York:Springer, 2011.
[9] RAZMERITA L. An ontology-based framework for modeling user behavior-a case study in knowledge management[J]. IEEE transactions on systems, man, and cybernetics-part A:systems and humans, 2011,41(4):772-783.
[10] 李学明,李海瑞,薛亮,等.基于信息增益与信息熵的TFIDF算法[J].计算机工程,2012,38(8):37-40.
[11] 王振振,何明,杜永萍.基于LDA主题模型的文本相似度计算[J].计算机科学,2013,40(12):229-232.
[12] 王传清,毕强. 超网络视域下的数字资源深度聚合研究[J]. 情报学报,2015(1):4-13.
[13] 刘洪伟,高鸿铭,陈丽,等.基于用户浏览行为的兴趣识别管理模型[J].数据分析与知识发现,2018(2):74-85.
[14] 曾子明,金鹏.基于用户兴趣变化的数字图书馆知识推荐服务研究[J].图书馆论坛,2016,36(1):94-99.
[15] KRISHNAMOORTHY R, SUNEETHA K R. User interest estimation using behavior monitoring measure[J]. Transplantation, 2013, 78(2):651-652.
[16] CLAYPOOL M, BROWN D, LE P, et al. Inferring user interest[J]. IEEE internet computing, 2001, 5(6):32-39.
[17] ZHENG L, CUI S, YUE D, et al. User interest modeling based on browsing behavior[C]//International conference on advanced computer theory and engineering. Chengdu:IEEE, 2010:V5-455-V5-458.
[18] 张海鹏. 基于Web日志挖掘的个性化推荐研究[D].重庆:重庆大学,2007.
[19] JEH G, WIDOM J. SimRank:a measure of structural-context similarity[C]//Eighth ACM SIGKDD international conference on knowledge discovery and data mining. Edmonton:ACM, 2002:538-543.
[20] 尹丽玲,刘柏嵩,王洋洋.跨类型的学术资源优质推荐算法研究[J].情报学报,2017,36(7):715-722.