期刊文献+

基于改进二部图与专家信任的混合推荐算法 被引量:2

Research of Hybrid Recommendation Algorithm Based on Improved Bipartite Network and Expert Trust
下载PDF
导出
摘要 针对传统协同过滤算法出现的稀疏数据、用户冷启动等问题以及复杂网络结构的广泛应用,本文提出结合改进的二部图与改进的专家信任算法来提高推荐准确度。基于普通二部图算法,将用户对项目的评分作为节点之间的分配资源权重,不仅关注用户与项目之间的联系,同时体现用户对项目的喜好程度;其次,本文根据用户的评论数和与该用户对项目评分相同的数目来判断该用户的专家信任度,改进传统系统过滤算法。为了提高推荐准确度,改进缺点,我们将两者算法进行加权混合,加权因子根据实验中最小MAE值对应的权值来确定,形成混合推荐算法。最后针对基于用户的协同过滤、传统二部图以及本文提出的混合算法计算MAE值和平均Hamming距离,对比分析本文算法的推荐准确度与多样性,实验表明本文方法推荐效果较好,准确率高,个性化强,有研究和应用价值。 In view of the limitations like data sparseness, new users with little record of traditional collaborative filtering recommendation algorithm and the wide application of complex network structure, this paper argues to coalesce the recommendation algorithm based on expert trust and the modified bipartite graph recommendation algorithm. First of all, we proposed an improved recommendation algorithm based on weighted networks, not only pay attention to the connection between users and projects, but also reflect the users' preferences in projects. Secondly, the degree of expert trust is determined by user comments and project reviews. And in order to improve the accuracy of recommendation, we coalesce this two kinds of algorithm and give them weight. Finally, we determine the weights and calculate MAE and the average Hamming distance of the traditional collaborative filtering, the traditional bipartite graph and the hybrid recommendation algorithm through experiments, which shows that the hybrid recommendation algorithm has higher accuracy, stronger individuation, and more research and application value.
出处 《价值工程》 2017年第19期160-164,共5页 Value Engineering
基金 大学生创新训练计划项目(编号:161005)
关键词 推荐算法 二部图 专家信任 加权混合 recommendation algorithm bipartite graph expert trust weighted mixture
  • 相关文献

参考文献2

二级参考文献27

  • 1WANG F H,JIAN S Y.An effective content-based recommendation method for Web browsing based on keyword context matching[J].Journal of Informatics and Electronics,2006,1 (2):49-59.
  • 2WARTENA C,SLAKHORST W,WIBBELS M,et al.Selecting keywords for content based recommendation[C] // CIKM'10:Proceedings of the 19th ACM International Conference on Information and Knowledge Management.New York:ACM Press,2010:1533-1536.
  • 3HERLOCKER J L,KONSTAN J A,TERVEEN L G,et al.Evaluating collaborative filtering recommender systems[J].ACM Transactions on Information Systems,2004,22(1):5-53.
  • 4CHEN Y L,CHENG L C.A novel collaborative filtering approach for recommending ranked items[J].Expert Systems with Applications,2008,34(4):2396-2405.
  • 5LI J,XU Y,WANG Y F,et al.Strongest association rules mining for efficient applications[C] //Proceedings of the Fourth IEEE Conference on Service Systems and Service Management.Piscataway,NJ:IEEE Press,2007:502-507.
  • 6WAND J C,CHIU C C.Recommending trusted online auction sellers using social network analysis[J].Expert Systems with Applications,2008,34(3):1666-1679.
  • 7ZHOU T,REN J,MEDO M,et al.Bipartite network projection and personal recommendation[J].Physical Review E,2007,76(4):6116-6123.
  • 8NEWMAN M E J.The structure and function of complex networks[J].SIAM Review,2003,45(2):167-256.
  • 9HOLME P,LILJEROS F,EDLING CR,et al.Network bipartivity[J].Physical Review E,2003,68(5):6108-6119.
  • 10LILJEROS F,EDLING C R,AMARAL L A N,et al.The Web of human sexual contacts[J].Nature,2001,411 (6840):907-908.

共引文献50

同被引文献26

引证文献2

二级引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部