面向冷启动用户的代表性物品选择
|
摘要
冷启动问题一直是推荐系统在实际应用过程中的一大难点,主动学习在推荐领域的应用一定程度上可以缓解这一困境.本文提出一个针对用户冷启动而生成"代表性物品"的主动学习策略.它利用用户与物品之间的关系,对用户与物品进行协同聚类,再借助于决策树得到最终的"代表性物品".实验证明,用"代表性物品"对用户进行分类后给出询问列表,能够获取到更多的评分数据以及更优的RMSE.
The cold start problem has always been a major issue in the practical application of the recommendation systems,and the application of active learning in the recommendation field has alleviated this dilemma to some degree. This paper proposes an active learning strategy for"representative items"generated for cold start users,utilizes the relationship between users and items to realize the collaborative clustering of users and items,and then obtains the final "representative items" with the help of decision tree models.Experiments show that by classifying users with "representative items" and giving them a personalized list of questions,more scoring data and better performance on RM SE metrics can be obtained.
The cold start problem has always been a major issue in the practical application of the recommendation systems,and the application of active learning in the recommendation field has alleviated this dilemma to some degree. This paper proposes an active learning strategy for"representative items"generated for cold start users,utilizes the relationship between users and items to realize the collaborative clustering of users and items,and then obtains the final "representative items" with the help of decision tree models.Experiments show that by classifying users with "representative items" and giving them a personalized list of questions,more scoring data and better performance on RM SE metrics can be obtained.
引文
[1]Bouguelia M R.An adaptive streaming active learning strategy based on instance w eighting[J].Pattern Recognition Letters,2016,70(C):38-44.
[2]Karimi R,Freudenthaler C,Nanopoulos A,et al.Active learning for aspect model in recommender systems[C]//Computational Intelligence and Data M ining,IEEE,2011:162-167.
[3]Hu Zhi-jie,Hu Yu-mo.Research on collaborative filtering recommendation bottleneck problem[J].Wireless Internet Technology,2016,9(46):100-101.
[4]Burr Settles.Active learning literature survey[C]//Computer Sciences Technical Report 1648,University of Wisconsin-M adison,2009:22-28.
[5]Harpale A S,Yang Y.Personalized active learning for collaborative filtering[C]//International ACM SIGIR Conference on Research and Development in Information Retrieval,ACM,2008:91-98.
[6]Wang L,Hu X,Yuan B,et al.Active learning via query synthesis and nearest neighbour search[J].Neurocomputing,2015,147(5):426-434.
[7]Barrett J E.Active learning for adaptive clinical trials:a streambased selective sampling strategy[J].Statistics,2015,36(5):62-69.
[8]Jones S,Shao L,Du K.Active learning for human action retrieval using query pool selection[J].Neurocomputing,2014,124(2):89-96.
[9]Sugiyama M,Nakajima S.Pool-based active learning in approximate linear regression[J].Machine Learning,2009,75(3):249-274.
[10]Elahi M,Ricci F,Rubens N.Active learning strategies for rating elicitation in collaborative filtering:a system-w ide perspective[J].Acm Transactions on Intelligent Systems&Technology,2014,5(1):1-33.
[11]Bridge D,Ricci F.Supporting product selection with query editing recommendations[C]//ACM Conference on Recommender Systems,ACM,2007:65-72.
[12]Kremer J,Steenstrup Pedersen K,Igel C.Active learning with support vector machines[J].Wiley Interdisciplinary Review s:Data M ining and Know ledge Discovery,2014,4(4):313-326.
[13]Nguyen H T,Smeulders A.Active learning using pre-clustering[C]//The Twenty-first International Conference on Machine Learning,ACM,2004:79-82.
[14]Huang A,Milne D,Frank E,et al.Clustering documents with active learning using w ikipedia[C]//Eighth IEEE International Conference on Data M ining,IEEE,2009:839-844.
[15]Golbandi N,Koren Y,Lempel R.Adaptive bootstrapping of recommender systems using decision trees[C]//Association for Computing M achinery,ACM,2011:595-604.
[16]Liu N N,Meng X,Liu C,et al.Wisdom of the better few:cold start recommendation via representative based rating elicitation[C]//ACM Conference on Recommender Systems,ACM,2011:37-44.
[17]Wu Dan,Yang Wei-dong.Cluster based keyword search of road netw ork[J].Journal of Chinese Computer Systems,2017,38(2):243-248.
[18]Wang Jun,Huang De-cai.Clustering method for position uncertain data based on connection number-ucndbscan[J].Journal of Chinese Computer Systems,2018,39(8):1633-1640.
[19]Carenini G,Smith J,Poole D.Towards more conversational and collaborative recommender systems[C]//International Conference on Intelligent User Interfaces,ACM,2003:12-18.
[20]Karimi R,Wistuba M,Nanopoulos A,et al.Factorized decision trees for active learning in recommender systems[C]//IEEE,International Conference on TOOLS w ith Artificial Intelligence,IEEE Computer Society,2013:404-411.
[21]Rasoul Karimi,Alexandros Nanopoulos,Lars Schmidt-Thieme.Improved questionnaire trees for active learning in recommender systems[J].Lernen,Wissen,Adaption,LWA,2014,8(10):1-11.
[22]Jain A K,Duin R P W,Mao J.Statistical pattern recognition:a review[J].IEEE Transactions on Pattern Analysis&M achine Intelligence,2002,22(1):4-37.
[23]Zong Yu,Jin Ping,Chen En-hong,et al.Fuzzy co-clustering algorithm for w eblog[J].Journal of Electronics&Information Technology,2012,34(3):543-548.
[24]Gao C F,Wu X J,Yu P.An Algorithm of fuzzy collaborative clustering based on kernel competitive agglomeration[J].Journal of Computers,2013,8(10):2623-2631.
[25]Zhu S,Yu K,Chi Y,et al.Combining content and link for classification using matrix factorization[C]//The 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval,ACM,2007:487-494.
[3]胡致杰,胡羽沫.协同过滤推荐瓶颈问题研究[J].无线互联科技,2016,9(46):100-101.
[17]吴丹,杨卫东.基于聚类的路网上关键字查询[J].小型微型计算机系统,2017,38(2):243-248.
[18]王骏,黄德才.一种新的位置不确定性聚类算法UCNDBSCAN[J].小型微型计算机系统,2018,39(8):1633-1640.
[23]宗瑜,金萍,陈恩红,等.面向Weblog的模糊协同聚类算法[J].电子与信息学报,2012,34(3):543-548.
[2]Karimi R,Freudenthaler C,Nanopoulos A,et al.Active learning for aspect model in recommender systems[C]//Computational Intelligence and Data M ining,IEEE,2011:162-167.
[3]Hu Zhi-jie,Hu Yu-mo.Research on collaborative filtering recommendation bottleneck problem[J].Wireless Internet Technology,2016,9(46):100-101.
[4]Burr Settles.Active learning literature survey[C]//Computer Sciences Technical Report 1648,University of Wisconsin-M adison,2009:22-28.
[5]Harpale A S,Yang Y.Personalized active learning for collaborative filtering[C]//International ACM SIGIR Conference on Research and Development in Information Retrieval,ACM,2008:91-98.
[6]Wang L,Hu X,Yuan B,et al.Active learning via query synthesis and nearest neighbour search[J].Neurocomputing,2015,147(5):426-434.
[7]Barrett J E.Active learning for adaptive clinical trials:a streambased selective sampling strategy[J].Statistics,2015,36(5):62-69.
[8]Jones S,Shao L,Du K.Active learning for human action retrieval using query pool selection[J].Neurocomputing,2014,124(2):89-96.
[9]Sugiyama M,Nakajima S.Pool-based active learning in approximate linear regression[J].Machine Learning,2009,75(3):249-274.
[10]Elahi M,Ricci F,Rubens N.Active learning strategies for rating elicitation in collaborative filtering:a system-w ide perspective[J].Acm Transactions on Intelligent Systems&Technology,2014,5(1):1-33.
[11]Bridge D,Ricci F.Supporting product selection with query editing recommendations[C]//ACM Conference on Recommender Systems,ACM,2007:65-72.
[12]Kremer J,Steenstrup Pedersen K,Igel C.Active learning with support vector machines[J].Wiley Interdisciplinary Review s:Data M ining and Know ledge Discovery,2014,4(4):313-326.
[13]Nguyen H T,Smeulders A.Active learning using pre-clustering[C]//The Twenty-first International Conference on Machine Learning,ACM,2004:79-82.
[14]Huang A,Milne D,Frank E,et al.Clustering documents with active learning using w ikipedia[C]//Eighth IEEE International Conference on Data M ining,IEEE,2009:839-844.
[15]Golbandi N,Koren Y,Lempel R.Adaptive bootstrapping of recommender systems using decision trees[C]//Association for Computing M achinery,ACM,2011:595-604.
[16]Liu N N,Meng X,Liu C,et al.Wisdom of the better few:cold start recommendation via representative based rating elicitation[C]//ACM Conference on Recommender Systems,ACM,2011:37-44.
[17]Wu Dan,Yang Wei-dong.Cluster based keyword search of road netw ork[J].Journal of Chinese Computer Systems,2017,38(2):243-248.
[18]Wang Jun,Huang De-cai.Clustering method for position uncertain data based on connection number-ucndbscan[J].Journal of Chinese Computer Systems,2018,39(8):1633-1640.
[19]Carenini G,Smith J,Poole D.Towards more conversational and collaborative recommender systems[C]//International Conference on Intelligent User Interfaces,ACM,2003:12-18.
[20]Karimi R,Wistuba M,Nanopoulos A,et al.Factorized decision trees for active learning in recommender systems[C]//IEEE,International Conference on TOOLS w ith Artificial Intelligence,IEEE Computer Society,2013:404-411.
[21]Rasoul Karimi,Alexandros Nanopoulos,Lars Schmidt-Thieme.Improved questionnaire trees for active learning in recommender systems[J].Lernen,Wissen,Adaption,LWA,2014,8(10):1-11.
[22]Jain A K,Duin R P W,Mao J.Statistical pattern recognition:a review[J].IEEE Transactions on Pattern Analysis&M achine Intelligence,2002,22(1):4-37.
[23]Zong Yu,Jin Ping,Chen En-hong,et al.Fuzzy co-clustering algorithm for w eblog[J].Journal of Electronics&Information Technology,2012,34(3):543-548.
[24]Gao C F,Wu X J,Yu P.An Algorithm of fuzzy collaborative clustering based on kernel competitive agglomeration[J].Journal of Computers,2013,8(10):2623-2631.
[25]Zhu S,Yu K,Chi Y,et al.Combining content and link for classification using matrix factorization[C]//The 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval,ACM,2007:487-494.
[3]胡致杰,胡羽沫.协同过滤推荐瓶颈问题研究[J].无线互联科技,2016,9(46):100-101.
[17]吴丹,杨卫东.基于聚类的路网上关键字查询[J].小型微型计算机系统,2017,38(2):243-248.
[18]王骏,黄德才.一种新的位置不确定性聚类算法UCNDBSCAN[J].小型微型计算机系统,2018,39(8):1633-1640.
[23]宗瑜,金萍,陈恩红,等.面向Weblog的模糊协同聚类算法[J].电子与信息学报,2012,34(3):543-548.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |