一种分段组合的个性化组推荐方法
|
摘要
提出一种分段组合两种推荐算法的后融合方法。首先选择两种基本算法对用户参与小组的历史信息进行挖掘,进而为用户推荐可能感兴趣的小组,然后对两种基本算法产生的推荐列表小组进行分段,接着利用所设计的分段函数组合预测概率,通过最远邻方法优化推荐小组,最后按照组合后的预测概率重新排序产生新的推荐列表。实验表明该方法提高了推荐的准确率,有利于用户更快速地定位感兴趣的小组。
Online community has become a popular social media platform,where users can share information,exchange opinions,and so on. Groups,as the core pattern of online community,have attracted a large amount of users. With the number of groups increasing,it is difficult for users to find out which group they may be interested in. Many researchers are paying attention to personalized group recommendation system to get over this problem. In this paper,we propose a novel framework for a group recommendation by integrating two different methods with a piecewise function. The framework contains four steps: dividing groups recommended by two different recommendation algorithms into three patterns,utilizing a piecewise function to compute the probability with which users will be interested in groups,filtering the groups by the furthest neighbor method,and reranking those groups. Experimental results on Douban show the effectiveness of our method by improving the accuracy of top-n recommdation. Finally,we develop a group recommendation system based on the proposed method.
Online community has become a popular social media platform,where users can share information,exchange opinions,and so on. Groups,as the core pattern of online community,have attracted a large amount of users. With the number of groups increasing,it is difficult for users to find out which group they may be interested in. Many researchers are paying attention to personalized group recommendation system to get over this problem. In this paper,we propose a novel framework for a group recommendation by integrating two different methods with a piecewise function. The framework contains four steps: dividing groups recommended by two different recommendation algorithms into three patterns,utilizing a piecewise function to compute the probability with which users will be interested in groups,filtering the groups by the furthest neighbor method,and reranking those groups. Experimental results on Douban show the effectiveness of our method by improving the accuracy of top-n recommdation. Finally,we develop a group recommendation system based on the proposed method.
引文
[1]艾瑞咨询.《2009年中国网络社区网民调研数据报告》[EB/OL].[2016-10-15].http://wenku.baidu.com/view/1fdef5f2 f90f76c661371a09.html.
[2]SCHAFER J B,FRANKOWSKI D,HERLOCKER J.Collaborative filtering recommender systems[J].Lect notes comput Sc,2007(4321):291-324.
[3]JONATHAN L.HERLOCKE R,JOSEPH A.An algorithmic framework for performing collaborative filtering[C]//SIGIR’99.2009:230-237.
[4]MUKUND deshpande,GEORGE Karypis.Item-based top-n recommendation algorithms[J].ACM transactions on information systems,2004,22(1):143-177.
[5]PAZZANI M J,BILLSUS D.Content-based recommendation systems[C]//The adaptive web.Springer Berlin Heidelberg,2007:325-341.
[6]LU Z,DOU Z,LIAN J,et al.Content-based collaborative filtering for news topic Recommendation[C]//AAAI.2015:217-223.
[7]王国霞,刘贺平.个性化推荐系统综述[J].计算机工程与应用,2012,48(7):16-19.
[8]WANG Jun,ARJEN P V,MARCEL J T.Reinders,unifying user-based and item-based collaborative filtering approaches by similarity fusion[C]//SIGIR’06.2010:501-508.
[9]MA Hao,IRWIN K,MICHAEL R L.Effective mssing data prediction for collaborative filtering[C]//SIGIR07.2013:39-46.
[10]汪静,印鉴,郑利荣,等.基于共同评分和相似权重的协同过滤推荐算法[J].计算机科学,2010,37(2):99-104.
[11]NORIAKI K.Serendipitous recommendations via innovators[C]//SIGIR’.2012,10:218-225.
[12]JAMALI M,ESTER M.A matrix factorization technique with trust propagation for recommendation in social networks[C]//Proceedings of the 4th ACM conference on recommender systems.2013:135-142.
[13]MA H,ZHOU D,LIU C.Recommender systems with social regularization[C]//Proceedings of the 4th ACM international conference on web search and data mining.2013:287-296.
[14]CHEN Wenyen,CHU Jonchyuan,LUAN Junyi.Collaborative filtering for orkut communities:discovery of user latent behavior[C]//World wide web.2009:681-690.
[15]ZHENG Nan,LI Qiudan,LIAO Shengcai.Which photo groups should I choose comparative study of recommendation algorithms in flickr[J].Journal of information science,2010,36(6):733-750.
[16]包红云,李秋丹,高珩,等.集成组内标签与用户链接关系的Flickr组推荐模型[J].模式识别与人工智能,2013,26(2):176-181.
[17]BADRUL Sarwar,GEORGE Karypis,JOSEPH Konstan,et al.Item-based collaborative filtering recommendation algorithms[C]//World wide web.2001:285-295.
[18]HERLICKE R,Konstan J L,Terveen J A.Evaluating collaborative filtering recommender systems[J].ACM transactions on information systems,2004(22):5-53.
[19]LIU Runran,JIA Chunxiao,ZHOU Tao,et al.Personal recommendation via modified collaborative filetring[J].Elsevier physica A,2009:462-468.
[20]MARK Claypool,ANUJA Gokhale,Tim Miranda.Combining content-based and collaborative filters in an online newspaper[C]//SIGIR’99,workshop recommendation systems:Algorithms and evaluation.2010.
[21]CHOI Sanghyun,JEONG Yongseon,Jeong K.A hybrid recommendation method with reduced data for largescale application[J].IEEE transacitions on systems,man,and cybernetics-part C:applications andreciews,2010,40(5):35-39.
[22]NEAL Lathia,STEPHEN Hailes,LICIA Capra,et al.Temporal diversity in recommender systems[C]//SIGIR’10,2012:210-217.
[23]ADOMAVICIUS G.Improving aggregate recommendation diversity using ranking-based techniques[J].IEEE transactions on knowledge and data engineering,2013,24(5):896-911.
[2]SCHAFER J B,FRANKOWSKI D,HERLOCKER J.Collaborative filtering recommender systems[J].Lect notes comput Sc,2007(4321):291-324.
[3]JONATHAN L.HERLOCKE R,JOSEPH A.An algorithmic framework for performing collaborative filtering[C]//SIGIR’99.2009:230-237.
[4]MUKUND deshpande,GEORGE Karypis.Item-based top-n recommendation algorithms[J].ACM transactions on information systems,2004,22(1):143-177.
[5]PAZZANI M J,BILLSUS D.Content-based recommendation systems[C]//The adaptive web.Springer Berlin Heidelberg,2007:325-341.
[6]LU Z,DOU Z,LIAN J,et al.Content-based collaborative filtering for news topic Recommendation[C]//AAAI.2015:217-223.
[7]王国霞,刘贺平.个性化推荐系统综述[J].计算机工程与应用,2012,48(7):16-19.
[8]WANG Jun,ARJEN P V,MARCEL J T.Reinders,unifying user-based and item-based collaborative filtering approaches by similarity fusion[C]//SIGIR’06.2010:501-508.
[9]MA Hao,IRWIN K,MICHAEL R L.Effective mssing data prediction for collaborative filtering[C]//SIGIR07.2013:39-46.
[10]汪静,印鉴,郑利荣,等.基于共同评分和相似权重的协同过滤推荐算法[J].计算机科学,2010,37(2):99-104.
[11]NORIAKI K.Serendipitous recommendations via innovators[C]//SIGIR’.2012,10:218-225.
[12]JAMALI M,ESTER M.A matrix factorization technique with trust propagation for recommendation in social networks[C]//Proceedings of the 4th ACM conference on recommender systems.2013:135-142.
[13]MA H,ZHOU D,LIU C.Recommender systems with social regularization[C]//Proceedings of the 4th ACM international conference on web search and data mining.2013:287-296.
[14]CHEN Wenyen,CHU Jonchyuan,LUAN Junyi.Collaborative filtering for orkut communities:discovery of user latent behavior[C]//World wide web.2009:681-690.
[15]ZHENG Nan,LI Qiudan,LIAO Shengcai.Which photo groups should I choose comparative study of recommendation algorithms in flickr[J].Journal of information science,2010,36(6):733-750.
[16]包红云,李秋丹,高珩,等.集成组内标签与用户链接关系的Flickr组推荐模型[J].模式识别与人工智能,2013,26(2):176-181.
[17]BADRUL Sarwar,GEORGE Karypis,JOSEPH Konstan,et al.Item-based collaborative filtering recommendation algorithms[C]//World wide web.2001:285-295.
[18]HERLICKE R,Konstan J L,Terveen J A.Evaluating collaborative filtering recommender systems[J].ACM transactions on information systems,2004(22):5-53.
[19]LIU Runran,JIA Chunxiao,ZHOU Tao,et al.Personal recommendation via modified collaborative filetring[J].Elsevier physica A,2009:462-468.
[20]MARK Claypool,ANUJA Gokhale,Tim Miranda.Combining content-based and collaborative filters in an online newspaper[C]//SIGIR’99,workshop recommendation systems:Algorithms and evaluation.2010.
[21]CHOI Sanghyun,JEONG Yongseon,Jeong K.A hybrid recommendation method with reduced data for largescale application[J].IEEE transacitions on systems,man,and cybernetics-part C:applications andreciews,2010,40(5):35-39.
[22]NEAL Lathia,STEPHEN Hailes,LICIA Capra,et al.Temporal diversity in recommender systems[C]//SIGIR’10,2012:210-217.
[23]ADOMAVICIUS G.Improving aggregate recommendation diversity using ranking-based techniques[J].IEEE transactions on knowledge and data engineering,2013,24(5):896-911.