[1]肖志宇,翟玉庆.改进的基于信任网络和随机游走策略的 评分预测模型[J].南京理工大学学报(自然科学版),2015,39(05):602.
 Xiao Zhiyu,Zhai Yuqing.Improved rating prediction model basing on trust network and random walk strategy[J].Journal of Nanjing University of Science and Technology,2015,39(05):602.
点击复制

改进的基于信任网络和随机游走策略的 评分预测模型
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
39卷
期数:
2015年05期
页码:
602
栏目:
出版日期:
2015-10-31

文章信息/Info

Title:
Improved rating prediction model basing on trust network and random walk strategy
作者:
肖志宇1翟玉庆12
东南大学 1.计算机科学与工程学院;
2.计算机网络和信息集成教育部重点实验室,江苏 南京 211189
Author(s):
Xiao Zhiyu1Zhai Yuqing12
1.School of Computer Science and Engineering;
2.Key Lab of Computer Network & Information Integration,Southeast University,Nanjing 211189,China
关键词:
推荐 信任网络 随机游走 评分预测 TrustWalker 用户相似度 TopN 评分参考用户
Keywords:
recommender trust network random walk rating prediction TrustWalker user similarity TopN rating referential users
分类号:
TP311
摘要:
为了提高推荐算法评分预测的准确度,该文在TrustWalker模型的基础上,提出了一个改进的基于信任网络和随机游走策略的评分预测模型——ReferentialUserWalker模型。该模型通过随机游走策略,利用信任网络中的信任朋友对目标物品或与目标物品相似的物品的评分进行评分预测,并在信任网络中找到最可信的TopN评分参考用户,同时引入信任度权重,降低了噪声数据的影响。实验结果表明,与TrustWalker模型相比,ReferentialUserWalker模型的评分预测准确度有所提高。
Abstract:
In order to improve the accuracy of the rating prediction in recommender systems,the ReferentialUserWalker model based on TrustWalker is proposed here.The model is combined with the random walk strategy on the trust network and the item-based recommendation to improve the accuracy of rating prediction and to find the TopN trusted rating referential users associated with the trust weight to predict the rating.And then effect of noise data is reduced.The experiment results prove that the model in this paper has higher accuracy of rating prediction than TrustWalker.

参考文献/References:

[1] Zan Huang,Zeng Daniel,Chen Hsinchun.A comparison of collaborative-filtering recommendation algorithms for e-commerce[J].IEEE Intelligent Systems,2007,22(5):68-78.
[2]项亮.推荐系统实践[M].北京:人民邮电出版社,2012.
[3]Massa P,Bhattacharjee B.Using trust in recommender systems:An experimental analysis[M].Berlin,Germany:Springer,2004.
[4]Golbeck J.Computing and applying trust in Web-based social networks[D].Washton D C,US:University of Maryland College Park,2005.
[5]Jamali M,Ester M.TrustWalker:A random walk model for combining trust-based and item-based recomm-endation[A].Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining[C].Pairs,France:ACM,2009:397-406.
[6]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms[A].Proceedings of the 10th International Conference on World Wide Web[C].Hong Kong,China:ACM,2001:285-295.
[7]Jannach D,Zanker M,Felfernig A,et al.Recommender systems:An introduction[M].New York,US:Cambridge University Press,2010.
[8]Watts D J.Six degrees:The science of a connected age[M].New York,US:WW Norton & Company,2004.
[9]Jamali S M.Probabilistic models for recommendation in social networks[D].Burnaby,Canada:School of Computing Science,Simon Fraser University,2013.
[10]Massa P.Epinions datasets[EB/OL].http://www.trustlet.org/wiki/Epinions_dataset,2011-11-04.

[11]Massa P,Avesani P.Trust-aware recommender systems[A].Proceedings of the 2007 ACM Conference on Recommender Systems[C].Minneapolis,US:ACM,2007:17-24.

[13]张富国.基于社交网络的个性化推荐技术[J].小型微型计算机系统,2014,35(7):1470-1476.
Zhang Fuguo.Survey of online social network based personalized recommendation[J].Journal of Chinese Computer Systems,2014,35(7):1470-1476.
[14]谭学清,黄翠翠,罗琳.社会化网络中信任推荐研究综述[J].现代图书情报技术,2014,30(11):10-16.
Tan Xueqing,Huang Cuicui,Luo Lin.A review of research on trust recommendation in social networks[J].New Technology of Library and Information Service,2014,30(11):10-16.

备注/Memo

备注/Memo:
收稿日期:2015-03-17 修回日期:2015-05-13
基金项目:国家自然科学基金(6509000059)
作者简介:肖志宇(1990-),男,硕士,主要研究方向:基于社交网络和信任模型的推荐系统,E-mail:xzy2502002@126.com; 通讯作者:翟玉庆(1966-),男,教授,主要研究方向:人工智能及其应用,E-mail:yqzhai@seu.edu.cn。
引文格式:肖志宇,翟玉庆.改进的基于信任网络和随机游走策略的评分预测模型[J].南京理工大学学报,2015,39(5):602-608.
投稿网址:http://zrxuebao.njust.edu.cn
DOI:10.14177/j.cnki.32-1397n.2015.39.05.015
更新日期/Last Update: 2015-10-31