计算机科学 ›› 2015, Vol. 42 ›› Issue (5): 34-41.doi: 10.11896/j.issn.1002-137X.2015.05.007
左万利,韩佳育,刘 露,王 英,彭 涛
ZUO Wan-li, HAN Jia-yu, LIU Lu, WANG Ying and PENG Tao
摘要: 了解用户兴趣是为用户提供个性化服务的关键。用户兴趣有短期兴趣和长期兴趣之分,且具有不稳定性。受人工免疫系统的启发,巧妙地将免疫应答过程应用于用户兴趣挖掘。首先将概率与时间相结合,提出“概念时序动态”的概念,以更好地刻画用户在一段时间内对同一兴趣的关注程度;然后基于人工免疫原理,建立抽取兴趣标签的分类器来提取用户兴趣标签;最后针对增量式学习,建立兴趣标签的“概念时序动态”,刻画出用户兴趣自首次出现以来受关注的程度,以此为依据来判断兴趣是否存在迁移及遗忘现象,并为每个兴趣标签附上权重。其主要贡献是创造性地将人工免疫原理应用于用户短期兴趣和长期兴趣的挖掘,并具有增量特性,可以很好地体现用户兴趣迁移特征,是一种自然完整的用户兴趣模型。实验结果表明,该学习模型能够很好地发现用户关注的领域,其平均精度和召回率分别达到79.5%和74.4%,是目前最贴近用户的兴趣挖掘模型。
[1] 林鸿飞,杨园生.用户兴趣模型的表示和更新机制[J].计算机研究与发展,2002,39(7):843-847 [2] Limam L,Coquil D,Kosch H.LionelBrunie:Extracting User Interests from Search Query Logs:A Clustering Approach[C]∥DEXA.2010:5-9 [3] Zhang Pin,Pu Ju-hua,Liu Yong-Li,et al.A probabilistic ap-proach for mining drifting user interests[C]∥APWeb/WAIM.2009:381-391 [4] 吴晶,张品,罗辛,等.门户个性化兴趣获取与迁移模式发现[J].计算机研究与发展,2007,4(8):1284-1292 [5] White R W,Bennett P N,Dumais S T.Predicting short-term interests using activity-based search context[C]∥CIKM.2010:1009-1018 [6] Holub M,Bieliková M.Estimation of user interest in visited web page[C]∥WWW.2010:1111-1112 [7] Cena F,Likavec S,Osborne F.Propagating User Interests inOntology-Based User Model[C]∥Proc.of XIIth International of Conference of the Italian Association for Artificial Intelligence.Berlin:Springer,2011:299-311 [8] Tan Bin,Lv Yuan-hua,Zhai Cheng-xiang.Mining long-lastingexploratory user interests from search history[C]∥CIKM.2012:1477-1481 [9] Chirita P-A,Nejdl W,Paiu R,et al.Using ODP Metadata to Personalize Search[C]∥SIGIR.2005:178-185 [10] So S,Lee J-H,Jung D,et al.Extending Open Directory Project to Represent User Interests[C]∥Proc.of 27th Annual ACM Symposium on Applied Computing.New York:ACM,2012:354-359 [11] Liu Tao,Zhou Yan,Hu Zhi-feng.A New Clustering AlgorithmBased on Artificial Immune System[C]∥Fuzzy Systems and Knowledge Discovery.IEEE,2008:347-351 [12] Watkins A B,Boggess L C.A Resource Limited Artificial Immune Classifier[C]∥Proceedings of the 2002 Congress on Evolutionary Computation Evolutionary Computation,2002(CEC ’02).2002:926-931 [13] Watkins A,Timmis J.Artificial Immune Recognition System(AIRS):An Immune-Inspired Supervised Learning Algorithm [J].Genetic Programming and Evolvable Machines,2004,5(3):291-317 [14] Greensmith J,Cayzer S.An Artificial Immune System Approach to Semantic Document Classification [C]∥Proc of Second International Conference on Artificial Immune Systems.LNCS 2787,Berlin:Springer,2003:136-146 [15] Manning C D,Raghavan P,Schutze H.信息检索导论[M].北京:人民邮电出版社,2010:81-83 [16] 莫宏伟,左兴权.人工免疫系统[M].北京:科学出版社,2009:96-100,0-410 [17] 韩家炜(加),堪博.数据挖掘概念与技术(第二版)[M].北京:机械工业出版社,2007:226-227 [18] Sugiyama K,Hatano K,Yoshikawa M.Adaptive Web Search Based on User Profile Constructed without Any Effort from U-sers[C]∥WWW’04.New York:ACM,2004:675-684 |
No related articles found! |
|