计算机科学 ›› 2016, Vol. 43 ›› Issue (2): 72-77.doi: 10.11896/j.issn.1002-137X.2016.02.016
• 2015年中国计算机学会人工智能会议 • 上一篇 下一篇
张东,蔡国永,夏彬彬
ZHANG Dong, CAI Guo-yong and XIA Bin-bin
摘要: 传统的推荐算法多以优化推荐列表的精确度为目标,而忽略了推荐算法的另一个重要指标——多样性。提出了一种新的提高推荐列表多样性的方法。该方法将列表生成步骤转换为N次概率选择过程,每次概率选择通过两个步骤完成:类型选择与项目选择。在类型选择中,引入项目的类型信息,根据用户对不同项目类型的喜好计算概率矩阵,并依照该概率矩阵选择一个类型;在项目选择中,根据项目的预测评分、项目的历史流行度、项目的推荐流行度3个因素重新计算项目的最终得分,选择得分最高的项目推荐给用户。通过阈值TR来调节多样性与精确度之间的折中。最后,通过对比实验证明了该方法的有效性。
[1] Cremonesi P,Koren Y,Turrin R.Performance of recommender algorithms on top-n recommendation tasks[C]∥Proc of the fourth ACM Conf on Recommender systems.ACM,2010:39-46 [2] McNee S M,Riedl J,Konstan J A.Being accurate is not enough:how accuracy metrics have hurt recommender systems[C]∥CHI’06 extended abstracts on Human factors in computing systems.ACM,2006:1097-1101 [3] Fleder D,Hosanagar K.Blockbuster culture’s next rise or fall:The impact of recommender systems on sales diversity[J].Mana-gement Science,2009,55(5):697-712 [4] Adamopoulos P,Tuzhilin A.On over-specialization and concentration bias of recommendations:Probabilistic neighborhood selection in collaborative filtering systems[C]∥Proc of the 8th ACM Conf on Recommender systems.ACM,2014:153-160 [5] Zhou T,Kuscsik Z,Liu J G,et al.Solving the apparent diversity-accuracy dilemma of recommender systems[J].Proc of the National Academy of Sciences,2010,107(10):4511-4515 [6] Park Y J,Tuzhilin A.The long tail of recommender systems and how to leverage it[C]∥Proc of the 2008 ACM conf on Recommender Systems.ACM,2008:11-18 [7] Yin H,Cui B,Li J,et al.Challenging the long tail recommendation[J].Proc of the VLDB Endowment,2012,5(9):896-907 [8] Ziegler C N,McNee S M,Konstan J A,et al.Improving recommendation lists through topic diversification[C]∥Proc of the 14th Int Conf on World Wide Web.ACM,2005:22-32 [9] Adomavicius G,Kwon Y O.Improving aggregate recommendation diversity using ranking-based techniques[J].IEEE Trans on Knowledge and Data Engineering,2012,24(5):896-911 [10] Vargas S,Castells P.Rank and relevance in novelty and diversity metrics for recommender systems[C]∥Proc of the fifth ACM Conf on Recommender Systems.ACM,2011:109-116 [11] Adomavicius G,Kwon Y.Maximizing aggregate recommenda-tion diversity:A graph-theoretic approach[C]∥Proc.of the 1st Int Workshop on Novelty and Diversity in Recommender Systems (DiveRS 2011).2011:3-10 [12] Vargas S,Castells P.Improving sales diversity by recommending users to items[C]∥Proc of the 8th ACM Conf on Recommender systems.ACM,2014:145-152 [13] Cremonesi P,Garzotto F,Negro S,et al.Looking for “good” re-commendations:A comparative evaluation of recommender systems[M]∥Human-Computer Interaction-INTERACT 2011.Springer Berlin Heidelberg,2011:152-168 [14] Bennett J,Lanning S.The netflix prize[C]∥Proc of KDD cup and workshop.2007:35 [15] Vargas S,Baltrunas L,Karatzoglou A,et al.Coverage,redundancy and size-awareness in genre diversity for recommender systems[C]∥Proc of the 8th ACM Conf on Recommender systems.ACM,2014:209-216 [16] Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems[J].Computer,2009,42(8):30-37 [17] Anderson C.The long tail:Why the future of business is selling less of more[M].Hyperion,2006 |
No related articles found! |
|