引入用户关注的图推荐模型的研究
|
摘要
随着互联网信息的快速增长,传统的检索技术已经无法满足用户快速及准确的访问信息的要求。现有的个性化应用推荐方法几乎都是依赖用户的搜索日志、点击日志或用户自身提供的标签来标识用户的兴趣(或称之为用户模型),然后依据此兴趣特征向用户推荐信息。本论文期望从用户间的特征相互关注这一角度入手,将用户的兴趣点予以全面挖掘,进而向用户推荐感兴趣的信息。基于此想法,本论文提出一种图推荐模型,该模型以用户和商品作为结点,并依据图中结点的分布情况计算每个结点的集中性以作为将某个项目推荐给用户的概率。该图推荐模型可以结合用户间的相互关注情况向用户推荐其感兴趣的商品,同时可以利用商品之间的相关性向用户推荐相关的商品。实验结果显示,该推荐模型无论在准确率上还是在运行效率上都优于传统的推荐模型。
With the fast development of Internet information, traditional retrieving techniques can no longer satisfy customers needs of quickly and precisely visiting desirable information. The present individualized recommendation methods mostly rely on the searching log, clicking log or the tags provided by users themselves to identify users' characteristics(or namely, user model). Then these characteristics are adopted to help guide recommendation. This paper hopes to start with the viewpoint of mutual following between users, which can fully explore the users' interests, and then recommend the information interested by users. Based on this idea, this paper proposes a graph based recommendation model. It combines users and items to form nodes, and calculate the centrality of each node via the distribution of nodes. The centrality is treated as the probability of recommending users in one node to the item in the same node. The proposed graph based recommendation model presented by this paper first considers the interests and preferences of the mutual following between users for recommending desirable product needed by users, and at the same time it can make use of the correlation between commodity recommend related products to users. Experimental results show that the proposed model is superior to the traditional recommendation model both in terms of accuracy and efficiency.
With the fast development of Internet information, traditional retrieving techniques can no longer satisfy customers needs of quickly and precisely visiting desirable information. The present individualized recommendation methods mostly rely on the searching log, clicking log or the tags provided by users themselves to identify users' characteristics(or namely, user model). Then these characteristics are adopted to help guide recommendation. This paper hopes to start with the viewpoint of mutual following between users, which can fully explore the users' interests, and then recommend the information interested by users. Based on this idea, this paper proposes a graph based recommendation model. It combines users and items to form nodes, and calculate the centrality of each node via the distribution of nodes. The centrality is treated as the probability of recommending users in one node to the item in the same node. The proposed graph based recommendation model presented by this paper first considers the interests and preferences of the mutual following between users for recommending desirable product needed by users, and at the same time it can make use of the correlation between commodity recommend related products to users. Experimental results show that the proposed model is superior to the traditional recommendation model both in terms of accuracy and efficiency.
引文
[1] 蔡强,韩东梅,李海生等.基于标签和协同过滤的个性化资源推荐[J].计算机科学,2014,41( 1):69-71.
[2] Guo L,Ma J,Chen Z,et al.Learning to recommend with social relation ensemble[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:2599-2602.
[3] Jiang M,Cui P,Liu R,et al.Social contextual recommendation[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:45-54.
[4] Sun J,Wang S,Gao B J,et al.Learning to rank for hybrid recommendation[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:2239-2242.
[5] Wang S,Sun J,Gao B J,et al.Adapting vector space model to ranking-based collaborative filtering[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:1487-1491.
[6] Ting Yuan,Jian Cheng,Xi Zhang,Qingshan Liu,Hanqing Lu.How friends affect user behaviors?An exploration of social relation analysis for recommendation[J].Knowledge-Based Systems,2015,88:70-84.
[7] Qinzhe Zhang,Jia Wu,Hong Yang,Weixue Lu,Guodong Long,Chengqi Zhang.Global and local influence-based social recommendation[C]// Proceedings of the 25th ACM International on Conference on Information and Knowledge Management,2016:1917-1920.
[8] Qiong Wu,Siyuan Liu,Chunyan Miao.Modeling uncertainty driven curiosity for social recommendation[C]//Proceedings of the International Conference on Web Intelligence,2017:790-798.
[9] Parisa Lak.A Novel approach to define and model contextual features in recommender systems[C]//Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval.ACM,2016:1161-1161.
[10] Nikolaos Pappas,Andrei Popescu-Belis.Combining content with user preferences for non-fiction multimedia recommendation:a study on TED lectures[J].Multimedia Tools and Applications,2015,74(4):1175-1197.
[11] Ibrahim Yakut,Jaideep Vaidya.Privacy-preserving item-based recommendations over partitioned data with overlaps[J].International Journal of Business Information Systems,2017,25(3):336-351.
[12] Vineeth Rakesh,Wang-Chien Lee,Chandan K.Reddy.Probabilistic group recommendation model for crowdfunding domains[C]//Proceedings of the Ninth ACM International Conference on Web Search and Data Mining.ACM,2016:257-266.
[13] Tony Cheng-Kui Huang,Yen-Liang Chen,Min-Chun Chen.A novel recommendation model with Google similarity[J].Decision Support Systems,2016,89:17-27.
[14] Zhijun Zhang,Hong Liu.Social recommendation model combining trust propagation and sequential behaviors[J].Applied Intelligence,2015,43(3):695-706.
[15] Qiudang Wang,Xiao Liu,Shasha Zhang,Yuanchun Jiang,Fei Du,Yading Yue,Yu Liang A Novel APP recommendation method based on SVD and social influence[C]//Proceedings,Part II,of the 15th International Conference on Algorithms and Architectures for Parallel Processing,2015:269-281.
[2] Guo L,Ma J,Chen Z,et al.Learning to recommend with social relation ensemble[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:2599-2602.
[3] Jiang M,Cui P,Liu R,et al.Social contextual recommendation[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:45-54.
[4] Sun J,Wang S,Gao B J,et al.Learning to rank for hybrid recommendation[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:2239-2242.
[5] Wang S,Sun J,Gao B J,et al.Adapting vector space model to ranking-based collaborative filtering[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.ACM,2012:1487-1491.
[6] Ting Yuan,Jian Cheng,Xi Zhang,Qingshan Liu,Hanqing Lu.How friends affect user behaviors?An exploration of social relation analysis for recommendation[J].Knowledge-Based Systems,2015,88:70-84.
[7] Qinzhe Zhang,Jia Wu,Hong Yang,Weixue Lu,Guodong Long,Chengqi Zhang.Global and local influence-based social recommendation[C]// Proceedings of the 25th ACM International on Conference on Information and Knowledge Management,2016:1917-1920.
[8] Qiong Wu,Siyuan Liu,Chunyan Miao.Modeling uncertainty driven curiosity for social recommendation[C]//Proceedings of the International Conference on Web Intelligence,2017:790-798.
[9] Parisa Lak.A Novel approach to define and model contextual features in recommender systems[C]//Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval.ACM,2016:1161-1161.
[10] Nikolaos Pappas,Andrei Popescu-Belis.Combining content with user preferences for non-fiction multimedia recommendation:a study on TED lectures[J].Multimedia Tools and Applications,2015,74(4):1175-1197.
[11] Ibrahim Yakut,Jaideep Vaidya.Privacy-preserving item-based recommendations over partitioned data with overlaps[J].International Journal of Business Information Systems,2017,25(3):336-351.
[12] Vineeth Rakesh,Wang-Chien Lee,Chandan K.Reddy.Probabilistic group recommendation model for crowdfunding domains[C]//Proceedings of the Ninth ACM International Conference on Web Search and Data Mining.ACM,2016:257-266.
[13] Tony Cheng-Kui Huang,Yen-Liang Chen,Min-Chun Chen.A novel recommendation model with Google similarity[J].Decision Support Systems,2016,89:17-27.
[14] Zhijun Zhang,Hong Liu.Social recommendation model combining trust propagation and sequential behaviors[J].Applied Intelligence,2015,43(3):695-706.
[15] Qiudang Wang,Xiao Liu,Shasha Zhang,Yuanchun Jiang,Fei Du,Yading Yue,Yu Liang A Novel APP recommendation method based on SVD and social influence[C]//Proceedings,Part II,of the 15th International Conference on Algorithms and Architectures for Parallel Processing,2015:269-281.