计算机科学 ›› 2019, Vol. 46 ›› Issue (11A): 153-158.

• 数据科学 • 上一篇    下一篇

主动学习在推荐系统中的应用

赵海燕1, 汪静1, 陈庆奎1, 曹健2   

  1. (上海理工大学光电信息与计算机工程学院 上海200093)1;
    (上海交通大学计算机科学与技术系 上海200030)2
  • 出版日期:2019-11-10 发布日期:2019-11-20
  • 通讯作者: 汪静(1994-),女,硕士生,主要研究方向为数据挖掘,E-mail:jingwang94@outlook.com。
  • 作者简介:赵海燕(1975-),女,博士,副教授,CCF会员,主要研究方向为个性化推荐、数据挖掘。
  • 基金资助:
    本文受国家自然科学基金项目(61272438,61202376,61472253),上海市科委项目(14511107702),上海市教委科研创新项目(13ZZ112,13YZ075)资助。

Application of Active Learning in Recommendation System

ZHAO Hai-yan1, WANG Jing1, CHEN Qing-kui1, CAO Jian2   

  1. (School of Optical-Electrical Information & Computer Engineering,University of Shanghai for Science & Technology,Shanghai 200093,China)1;
    (Department of Computer Science & Technology,Shanghai Jiao Tong University,Shanghai 200030,China)2
  • Online:2019-11-10 Published:2019-11-20

摘要: 近年来,推荐技术迅速发展,日趋成熟。但是,多数推荐算法都建立在一个理想的假设下,即有足够多的样本数据供我们训练出成熟的模型用于预测或推荐。在实际工业化生产中,一方面,大多数的用户和项目只拥有极少量的标签信息;另一方面,即使依靠历史积累形成的数据集,在分布上也十分不均衡,难以学习出可靠的推荐模型。主动学习的思想认为每个项目给系统带来的“好处”是不等的,因而可以通过特定策略选择某些项目,借助用户与项目之间的交互行为来主动获取相关的偏好信息。应用在推荐系统中的主动学习试图选择数量更少、质量更高的样本来训练模型,既能提高用户体验,又能免受数据集不均衡的束缚。文中综述了近年来主动学习在推荐系统中的应用,并对其发展趋势进行分析。

关键词: 成员查询, 基于会话, 冷启动, 推荐系统, 主动学习

Abstract: In recent years,recommender system develops very quickly and is becoming more and more mature.However,many approaches are based on an ideal assumption,i.e.,there are plenty of sample data which can help us train a mature model to predict or recommend.In actual industrial production,most users and products lack of rating information or consumption records.And datasets formed by historical accumulation are unevenly distribued,so that it is hard to learn a reliable model.Active learning considers that the benefits of each item to the system is different,so some special items can be selected through specific strategies,and the related preference information can be actively obtained by the interaction between the user and the project.Active learning applied in the recommendation system attempts to training a model with fewer but higher quality samples,which improves the user experience and protects against unbalanced data sets.The applications of active learning in recommendation system in recent years were reviewed and summarized.Future directions were also discussed in this paper.

Key words: Active learning, Cold start, Conversational-based, Query-by-committ, Recommendation system

中图分类号: 

  • TP311
[1]CELMAÒ.Music Recommendation and Discovery in the LongTail[J].Ceedings of International Congress on Electron Microscopy Methods Enzymol-89,2008,11(1):7-8.
[2]KARIMI R,WISTUBA M,NANOPOULOS A,et al.Factorized Decision Trees for Active Learning in Recommender Systems[C]∥IEEE International Conference on TOOLS with Artificial Intelligence.IEEE Computer Society,2013:404-411.
[3]ELAHI M,RICCI F,RUBENS N.Active Learning in Collabora-tive Filtering Recommender Systems[J].Computer Science Review,2016,20(C):29-50.
[4]POZO M,CHIKY R,MEZIANE F,et al.Enhancing new user cold-start based on decision trees active learning by using past warm-users predictions[J].International Conference on Computational Collective Intelligence,2017,7(4):137-147.
[5]BOUTILIER C,ZEMEL R S,MARLIN B.Active Collaborative Filtering[C]∥19th Conference on Uncertainty in Artifical Intelligence.2012:98-106.
[6]HOFMANN T.Latent semantic models for collaborative filte-ring[J].ACM Transactions on Information Systems,2013,22(1):89-115.
[7]SETTLES B.Active Learning Literature Survey:ComputerScience Technical Report 1648[R].University of Wisconsin,Madison,2010.
[8]CASSEL S,HOWAR F,JONSSON B,et al.Active learning for extended finite state machines[J].Formal Aspects of Computing,2016,28(2):233-263.
[9]SHARMA M,BILGIC M.Evidence-based uncertainty sampling for active learning[J].Data Mining & Knowledge Discovery,2016,31(1):1-39.
[10]HUANG E C,PAO H K,LEE Y J.Big active learning[C]∥IEEE International Conference on Big Data.IEEE,2018:94-101.
[11]NARR A,TRIEBEL R,CREMERS D.Stream-based ActiveLearning for efficient and adaptive classification of 3D objects[C]∥IEEE International Conference on Robotics and Automation.IEEE,2016:227-233.
[12]LANG K J,BAUM E B.Query learning can work poorly when a human oracle is used[C]∥IEEE Intl. Joint Conference on Neural Networks.1992:306-322.
[13]HOI S C H,JIN R,ZHU J,et al.Semi-supervised SVM batch mode active learning for image retrieval[C]∥IEEE Conference on Computer Vision and Pattern Recognition,2008(CVPR 2008).IEEE,2008:1-7.
[14]XU Z,YU K,TRESP V,et al.Representative Sampling for Text Classification Using Support Vector Machines[C]∥European Conference on Ir Research.Springer-Verlag,2003:393-407.
[15]JONES S,SHAO L,DU K.Active learning for human action retrieval using query pool selection[J].Neurocomputing,2014,124(2):89-96.
[16]SUGIYAMA M,NAKAJIMA S.Pool-based active learning inapproximate linear regression[J].Machine Learning,2009,75(3):249-274.
[17]BOUGUELIA M R.An adaptive streaming active learning strategy based on instance weighting[J].Pattern Recognition Letters,2016,70:38-44.
[18]BOUGUELIA M R,BELAÏD Y,BELAÏD A.A Stream-Based Semi-supervised Active Learning Approach for Document Classification[C]∥International Conference on Document Analysis and Recognition.IEEE,2013:611-615.
[19]ROKACH L,NAAMANI L,SHMILOVICI A.Pessimistic cost-sensitive active learning of decision trees for profit maximizing targeting campaigns[J].Data Mining & Knowledge Discovery,2008,17(2):283-316.
[20]KOHRS A,BERNARD M.Improving collaborative filtering for new users by smart object selection[C]∥Proceedings of International Conference on Media Features.2001.
[21]HOFMANN T.Collaborative filtering via gaussian probabilistic latent semantic analysis[C]∥ACM SIGIR Conference.2003:259-266.
[22]SCHOHN G,COHN D.Less is More:Active Learning withSupport Vector Machines[C]∥Seventeenth International Conference on Machine Learning.Morgan Kaufmann Publishers Inc.,2000:839-846.
[23]SETTLES B,CRAVEN M,RAY S.Multiple-Instance ActiveLearning[C]∥Conference on Neural Information Processing Systems.DBLP,2008:1289-1296.
[24]TONG S,KOLLER D.Support vector machine active learning with applications to text classification[J].Journal of Machine Learning Research,2001,2(1):999-1006.
[25]JAN K P S,CHRISTIAN K I.Active learning with support vector machines[J].Wiley Interdisciplinary Reviews:Data Mining and Knowledge Discovery,2004,4(4):313-326.
[26]ROY N,MCCALLUM A.Toward optimal active learningthrough monte carlo estimation of error reduction[C]∥ICML.2001:441-448.
[27]GEMAN S,BIENENSTOCK E.Neural networks and the bias/variance dilemma[M].MIT Press,1992:1-58.
[28]GUAN C,LIU Q,LV J,et al.Consolidation:Metric+ActiveLearning and Its Applications for Cross-Domain Recommendation[C]∥IEEE/Wic/ACM International Conference on Web Intelligence and Intelligent Agent Technology.IEEE,2016:244-251.
[29]SUGIYAMA M.Active Learning in Approximately Linear Regression Based on Conditional Expectation of Generalization Error[J].Journal of Machine Learning Research,2006,7(1):141-166.
[30]TEIXEIRA I R,CARVALHO F D A T D,RAMALHO G L,et al.ActiveCP:A Method for Speeding up User Preferences Acquisition in Collaborative Filtering Systems[M]∥Advances in Artificial Intelligence.Springer Berlin Heidelberg,2002:237-247.
[31]GOLBANDI N,KOREN Y,LEMPEL R.On bootstrapping re-commender systems[C]∥ACM Internatioanl Conference on Information & Knowledge Management.ACM,2010:1805-1808.
[32]ZHAO Y,XU C,CAO Y.Research on query-by-committeemethod of active learning and application[C]∥International Conference on Advanced Data Mining and Applications.Sprin-ger-Verlag,2016:985-991.
[33]ELAHI M,RICCI F,RUBENS N.Active learning strategies for rating elicitation in collaborative filtering:A system-wide perspective[J].Acm Transactions on Intelligent Systems & Technology,2014,5(1):1-33.
[34]ELAHI M,REPSYS V,RICCI F.Rating Elicitation Strategies for Collaborative Filtering[C]∥E-Commerce and Web Techno-logies.2011,85:160-171.
[35]RASHID A M,ALBERT I,DAN C,et al.Getting to know you:learning new user preferences in recommender systems[C]∥International Conference on Intelligent User Interfaces.ACM,2002:127-134.
[36]PAGANO R,QUADRANA M,ELAHI M,et al.Toward Active Learning in Cross-domain Recommender Systems[J].arXiv:1701.02021,2017.
[37]KOREN Y,BELL R.Advances in Collaborative Filtering[M]∥Recommender Systems Handbook.2011:145-186.
[38]ELAHI M,BRAUNHOFER M,RICCI F,et al.Personality-Based Active Learning for Collaborative Filtering Recommender Systems[C]∥Congress of the Italian Association for Artificial Intelligence.Cham:Springer,2013:360-371.
[39]SEUNG H,OPPER S.Query by committee[J].Proc of the Fith Workshop on Computational Learning Theory,1992,284:287-294.
[40]GEIGER D.Current State of Personalized Task Recommenda-tion[M]∥Personalized Task Recommendation in Crowdsourcing Systems.Springer International Publishing,2016:15-29.
[41]KRAWCZYK B,WOÒNIAK M.Online query by committee for active learning from drifting data streams[C]∥International Joint Conference on Neural Networks.IEEE,2017:38-49.
[42]PICHARA K,SOTO A,ARANEDA A.Detection of Anomalies in Large Datasets Using an Active Learning Scheme Based on Dirichlet Distributions[C]∥Ibero-American Conference on Ai:Advances in Artificial Intelligence.Springer-Verlag,2008:163-172.
[43]NGUYEN T T,PHAM H V,VU P M,et al.RecommendingAPI Usages for Mobile Apps with Hidden Markov Model[C]∥IEEE/ACM International Conference on Automated Software Engineering.IEEE,2016:795-800.
[44]SHI S,LIU Y,HUANG Y,et al.Active Learning for kNNBased on Bagging Features[C]∥International Conference on Natural Computation.IEEE,2008:61-64.
[45]FATHIAN M,HOSEINPOOR Y,MINAEI-BIDGOLI B.Offering a hybrid approach of data mining to predict the customer churn based on bagging and boosting methods[J].Kybernetes,2016,45(5):732-743.
[46]ZHAO Y,XU C,CAO Y.Research on query-by-committeemethod of active learning and application[C]∥International Conference on Advanced Data Mining and Applications.Sprin-ger-Verlag,2006:985-991.
[47]RASHID A M,KARYPIS G,RIEDL J.Learning preferences of new users in recommender systems:an information theoretic approach[J].ACM Sigkdd Explorations Newsletter,2008,10(2):90-100.
[48]CARENINI G,SMITH J,POOLE D.Towards more conversational and collaborative recommender systems[C]∥InternationalConference on Intelligent User Interfaces.ACM,2003:12-18.
[49]PENNOCK D M,HORVITZ E,LAWRENCE S,et al.Collaborative Filtering by Personality Diagnosis:A Hybrid Memory-and Model-Based Approach[C]∥Proceedings of IJCAI-99.2013:473-480.
[50]SETTLES B,CRAVEN M.An analysis of active learning strategies for sequence labeling tasks[C]∥Conference on Empirical Methods in Natural Language Processing.Association for Computational Linguistics,2008:1070-1079.
[51]IKHWANTRI F,LOUVAN S,KURNIAWAN K,et al.Multi-Task Active Learning for Neural Semantic Role Labeling on Low Resource Conversational Corpus[C]∥Proceedings of the 2007 ACM Conference on Recommender Systems.2018:37-49.
[52]BRIDGE D,RICCI F.Supporting product selection with query editing recommendations[C]∥ACM Conference on Recommender Systems.ACM,2007:65-72.
[53]RICCI F,NGUYEN Q N.Acquiring and Revising Preferences in a Critique-Based Mobile Recommender System[J].IEEE Intelligent Systems,2007,22(3):22-29.
[54]MCNEE S M,LAM S K,KONSTAN J A,et al.Interfaces for eliciting new user preferences in recommender systems[C]∥International Conference on User Modeling.Springer-Verlag,2003:178-187.
[55]GOLBANDI N,KOREN Y,LEMPEL R.Adaptive bootstrap-ping of recommender systems using decision trees[C]∥ACM International Conference on Web Search and Data Mining.ACM,2011:595-604.
[56]KARIMI R,WISTUBA M,NANOPOULOS A,et al.Factorized Decision Trees for Active Learning in Recommender Systems[C]∥IEEE,International Conference on TOOLS with Artificial Intelligence.IEEE Computer Society,2013:404-411.
[57]HERLOCKER J L,KONSTAN J A,TERVEEN L G,et al.Evaluating collaborative filtering recommender systems[J].ACM Transactions on Information Systems,2004,22(1):5-53.
[58]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.
[59]SHANI G,GUNAWARDANA A.Evaluating Recommendation Systems[M]∥Recommender Systems Handbook.Springer,2011:257-297.
[60]LIU N N,YANG Q.EigenRank:a ranking-oriented approach to collaborative filtering[C]∥International ACM SIGIR Confe-rence on Research and Development in Information Retrieval.ACM,2008:83-90.
[1] 程章桃, 钟婷, 张晟铭, 周帆.
基于图学习的推荐系统研究综述
Survey of Recommender Systems Based on Graph Learning
计算机科学, 2022, 49(9): 1-13. https://doi.org/10.11896/jsjkx.210900072
[2] 张佳, 董守斌.
基于评论方面级用户偏好迁移的跨领域推荐算法
Cross-domain Recommendation Based on Review Aspect-level User Preference Transfer
计算机科学, 2022, 49(9): 41-47. https://doi.org/10.11896/jsjkx.220200131
[3] 王冠宇, 钟婷, 冯宇, 周帆.
基于矢量量化编码的协同过滤推荐方法
Collaborative Filtering Recommendation Method Based on Vector Quantization Coding
计算机科学, 2022, 49(9): 48-54. https://doi.org/10.11896/jsjkx.210700109
[4] 周芳泉, 成卫青.
基于全局增强图神经网络的序列推荐
Sequence Recommendation Based on Global Enhanced Graph Neural Network
计算机科学, 2022, 49(9): 55-63. https://doi.org/10.11896/jsjkx.210700085
[5] 秦琪琦, 张月琴, 王润泽, 张泽华.
基于知识图谱的层次粒化推荐方法
Hierarchical Granulation Recommendation Method Based on Knowledge Graph
计算机科学, 2022, 49(8): 64-69. https://doi.org/10.11896/jsjkx.210600111
[6] 方义秋, 张震坤, 葛君伟.
基于自注意力机制和迁移学习的跨领域推荐算法
Cross-domain Recommendation Algorithm Based on Self-attention Mechanism and Transfer Learning
计算机科学, 2022, 49(8): 70-77. https://doi.org/10.11896/jsjkx.210600011
[7] 周慧, 施皓晨, 屠要峰, 黄圣君.
基于主动采样的深度鲁棒神经网络学习
Robust Deep Neural Network Learning Based on Active Sampling
计算机科学, 2022, 49(7): 164-169. https://doi.org/10.11896/jsjkx.210600044
[8] 帅剑波, 王金策, 黄飞虎, 彭舰.
基于神经架构搜索的点击率预测模型
Click-Through Rate Prediction Model Based on Neural Architecture Search
计算机科学, 2022, 49(7): 10-17. https://doi.org/10.11896/jsjkx.210600009
[9] 齐秀秀, 王佳昊, 李文雄, 周帆.
基于概率元学习的矩阵补全预测融合算法
Fusion Algorithm for Matrix Completion Prediction Based on Probabilistic Meta-learning
计算机科学, 2022, 49(7): 18-24. https://doi.org/10.11896/jsjkx.210600126
[10] 侯夏晔, 陈海燕, 张兵, 袁立罡, 贾亦真.
一种基于支持向量机的主动度量学习算法
Active Metric Learning Based on Support Vector Machines
计算机科学, 2022, 49(6A): 113-118. https://doi.org/10.11896/jsjkx.210500034
[11] 蔡晓娟, 谭文安.
一种改进的融合相似度和信任度的协同过滤算法
Improved Collaborative Filtering Algorithm Combining Similarity and Trust
计算机科学, 2022, 49(6A): 238-241. https://doi.org/10.11896/jsjkx.210400088
[12] 何亦琛, 毛宜军, 谢贤芬, 古万荣.
基于点割集图分割的矩阵变换与分解的推荐算法
Matrix Transformation and Factorization Based on Graph Partitioning by Vertex Separator for Recommendation
计算机科学, 2022, 49(6A): 272-279. https://doi.org/10.11896/jsjkx.210600159
[13] 郭亮, 杨兴耀, 于炯, 韩晨, 黄仲浩.
基于注意力机制和门控网络相结合的混合推荐系统
Hybrid Recommender System Based on Attention Mechanisms and Gating Network
计算机科学, 2022, 49(6): 158-164. https://doi.org/10.11896/jsjkx.210500013
[14] 熊中敏, 舒贵文, 郭怀宇.
融合用户偏好的图神经网络推荐模型
Graph Neural Network Recommendation Model Integrating User Preferences
计算机科学, 2022, 49(6): 165-171. https://doi.org/10.11896/jsjkx.210400276
[15] 洪志理, 赖俊, 曹雷, 陈希亮, 徐志雄.
基于遗憾探索的竞争网络强化学习智能推荐方法研究
Study on Intelligent Recommendation Method of Dueling Network Reinforcement Learning Based on Regret Exploration
计算机科学, 2022, 49(6): 149-157. https://doi.org/10.11896/jsjkx.210600226
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!