《计算机应用研究》|Application Research of Computers

考虑社会关系影响差异和动态性的社会化推荐

Social recommendation based on considerations of different effects and dynamicity of social relations

免费全文下载 (已被下载 次)  
获取PDF全文
作者 赵海燕,韩松,陈庆奎,曹健
机构 1.上海理工大学 光电信息与计算机工程学院 上海现代光学系统重点实验室,上海 200093;2.上海交通大学 计算机科学与技术系,上海 200030
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2018)09-2605-05
DOI 10.3969/j.issn.1001-3695.2018.09.010
摘要 随着社交媒体的发展,社交活动日益普及并产生丰富的社会关系。融合社会关系进行推荐可以缓解传统推荐系统面临的数据稀疏和冷启动问题。已有很多成功利用社会关系和评分信息进行推荐的算法,然而它们没有充分地挖掘不同的社会关系对用户的不同影响以及在不同时间段中社会关系的不同影响,这导致推荐效果的下降。基于对现实中社会关系影响复杂性的充分考虑,提出了新的考虑社会关系影响差异和动态性的社会化推荐算法。在Epinions数据集上的实验结果表明该方法可以提高推荐质量。复杂度分析也表明该方法具有可扩展性,能够适应大数据集的要求。
关键词 协同过滤;社交网络;矩阵分解;社交信任;差异性;动态性
基金项目 国家自然科学基金资助项目(61272438,61202376,61472253)
上海市科委资助项目(14511107702)
上海市教委科研创新项目(13ZZ112
13YZ075)
本文URL http://www.arocmag.com/article/01-2018-09-010.html
英文标题 Social recommendation based on considerations of different effects and dynamicity of social relations
作者英文名 Zhao Haiyan, Han Song, Chen Qingkui, Cao Jian
机构英文名 1.ShanghaiKeyLaboratoryofModernOpticalSystem,SchoolofOpticalElectrical&ComputerEngineering,UniversityofShanghaiforScience&Technology,Shanghai200093,China;2.Dept.ofComputerScience&Technology,ShanghaiJiaoTongUniversity,Shanghai200030,China
英文摘要 With the development of social media, social activities are more and more popular among people and rich social relations have been produced.Fusing social relations to recommend can potentially alleviate data sparse and cold start problems that traditional recommender systems face.Many successful recommendation algorithms employing both social relations and rating information have been proposed in recent years.However, these existing algorithms do not consider much on the differences of the influences generated by each social relation and the temporal changes of social factors, which degraded the performance of recommendation.Based on the fact that the social relations in the real world are very complex, this paper proposed new social recommendation algorithms that considered the different effects and dynamicity of social relations.The experimental results on Epinions dataset show that this method can improve the recommendation quality.The complexity analysis also indicates that this approach can be applied to large datasets.
英文关键词 collaborative filtering; social network; matrix factorization; social trust; difference; dynamicity
参考文献 查看稿件参考文献
  [1] Sarwar B, Karypis G, Konstan J, et al. Item-based collaborative filtering recommendation algorithms[C] //Proc of the 10th International Conference on World Wide Web. New York:ACM Press, 2001:285-295.
[2] Golbeck J A. Computing and applying trust in Web-based social networks[D] . Maryland:University of Maryland at College Park, 2005.
[3] Massa P, Avesani P. TrustAware recommender systems[C] //Proc of ACM Conference on Recommender Systems. New York:ACM Press, 2007:17-24.
[4] Ma Hao, Yang Haixuan, Lyu M R, et al. Sorec:social recommendation using probabilistic matrix factorization[C] //Proc of the 17th ACM Conference on Information and Knowledge Management. New York:ACM Press, 2008:931-940.
[5] Jamali M, Ester M. A matrix factorization technique with trust propagation for recommendation in social networks[C] //Proc of the 4th ACM Conference on Recommender Systems. New York:ACM Press, 2010:135-142.
[6] Ma Hao, Zhou Dengyang, Liu Chao, et al. Recommender systems with social regularization[C] //Proc of the 4th ACM International Conference on Web Search and Data Mining. New York:ACM Press, 2011:287-296.
[7] Tang Jiliang, Hu Xia, Gao Huiji, et al. Exploiting local and global social context for recommendation[C] //Proc of the 23rd International Joint Conference on Artificial Intelligence. Polo Alto, CA:AAAI Press, 2013:264-269.
[8] Xiang Rongjing, Neville J, Rogati M. Modeling relationship strength in online social networks[C] //Proc of the 19th International Conference on World Wide Web. New York:ACM Press, 2010:981-990.
[9] McPherson M, Smith-Lovin L, Cook J M. Birds of a feather:homophily in social networks[J] . Annual Review of Sociology, 2001, 27(1):415-444.
[10] Marsden P V, Friedkin N E. Network studies of social influence[J] . Sociological Methods & Research, 1993, 22(1):127-151.
[11] Koren Y. Collaborative filtering with temporal dynamics[J] . Communications of the ACM, 2010, 53(4):89-97.
[12] 项亮. 推荐系统实践[M] . 北 京:人 民 邮 电 出 版 社, 2012:44-64.
[13] Liu Haifeng, Hu Zheng, Mian A, et al. A new user similarity model to improve the accuracy of collaborative filtering[J] . Knowledge-Based Systems, 2014, 56(1):156-166.
[14] Tang Jiliang, Gao Huiji, Hu Xia, et al. Context-aware review helpfulness rating prediction[C] //Proc of the 7th ACM Conference on Recommender Systems. New York:ACM Press, 2013:1-8.
[15] Victor P, Cornelis C, De Cock M, et al. A comparative analysis of trust-enhanced recommenders for controversial items[C] //Proc of the 3rd International Conference on Weblogs and Social Media. [S. l. ] :ICWSM Press, 2009:342-345.
[16] Victor P, De Cock M, Cornelis C. Trust and recommendations[M] //Recommender Systems Handbook. Berlin:Springer, 2011:645-675.
[17] Jamali M, Ester M. TrustWalker:a random walk model for combining trust-based and item-based recommendation[C] //Proc of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 2009:397-406.
[18] Mnih A, Salakhutdinov R. Probabilistic matrix factorization[C] //Advances in Neural Information Processing Systems. 2008:1257-1264.
[19] Luo Xin, Zhou Mengchu, Xia Yunni, et al. An efficient non-negative matrix-factorization-based approach to collaborative filtering for recommender systems[J] . IEEE Trans on Industrial Informatics, 2014, 10(2):1273-1284.
[20] Ma Hao, King I, Lyu M R. Learning to recommend with social trust ensemble[C] //Proc of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM Press, 2009:203-210.
收稿日期 2017/4/19
修回日期 2017/5/31
页码 2605-2609
中图分类号 TP391
文献标志码 A