计算机科学 ›› 2014, Vol. 41 ›› Issue (8): 7-12.doi: 10.11896/j.issn.1002-137X.2014.08.002
郑志高,刘京,王平,孙圣力
ZHENG Zhi-gao,LIU Jing,WANG Ping and SUN Sheng-li
摘要: 围绕传统的协同过滤推荐算法存在的局限性展开研究,提出一种时间加权不确定近邻协同过滤推荐算法TWUNCF。根据推荐系统应用的实际情况,首先对用户和产品相似度进行时间加权以保证数据有效性,在此基础上改进相似度的计算方法。同时引入近邻因子在产品群和用户群中自适应地选择预测目标的近邻对象作为推荐群,计算推荐群中推荐概率较高的信任子群,最后通过不确定近邻的动态度量方法来对预测结果进行平衡的推荐。实验结果表明,该算法考虑了数据的时间有效性,同时平衡不同群体对推荐结果的影响,避免由于数据稀疏带来的推荐结果不准确和计算难度大的问题。理论分析和模拟实验证明,该算法在一定程度上提高了系统的准确性和推荐效率。
[1] Jelenc David.Decision making matters:A better way to evaluate trust models [J].Knowledge-Based Systems,2013,2:147-164 [2] Chen M-C,Chen L-S,Hsu F-H,et al.HPRS:A profitabilitybased recommender system [C]∥Helander M,Xie Min,Jiao R,et al,eds.Proceeding of the IEEE International Conference on Industrial Engineering and Engineering Management.Singapore,IEEE Engineering Management Society Singapore Chapter,2007(12):219-223 [3] Xu Xiao-wei,Wang Fu-dong.Trust -Based Collaborative Fil-ter- ing Algorithm [C]∥Proceedings of the 2012 Fifth Interna-tional Symposium on Computational Intelligence and Design.Washington DC,USA,2012:321-324 [4] Park S-T,Pennock D M.Applying collaborative filtering techniques to movie search for better ranking and browsing [C]∥Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,KDD 2007.San Jose,California,USA, August 2007:834-843 [5] Tomoharu I,Kazumi S,Takeshi Y.Modeling user behavior inrecommender system based on maximum entropy [C]∥Proceedings of the 16th International Conference on World Wide Web. Canada:World Scientific Publishing Co.Pte.Ltd.,2007:1281-1282 [6] Pappageorge,Charlotte.Recommended verification approach for human rated systems based on lessons learned on the orion launch abort system [C]∥21st Annual International Symposium of the International Council on Systems Engineering.2011,1:1-13 [7] Kaklauskas A.Recommended biometric stress management system [J].Expert Systems with Applications,2011,38(11):14011-14025 [8] Massa P,Avesani P.Trust-aware collaborative filtering for recommender systems [J].Lecture Notes in Computer Science,2004,3290:492-508 [9] Vincent S-Z,Boi Faltings.Using hierarchical clustering forlearning the ontologies used in recommendation systems [C]∥Proceedings of the 13th ACM SIGKDD International Confe-rence on Knowledge Discovery and Data Mining.San Jose,California,United States,2007:599-608 [10] Yang J-M,Li K-F.An interference-based collaborative fil-tering approach [C]∥Proceedings of the 3rd IEEE International Symposium on Dependable,Autonomic and Secure Computing (DASC).Columbia,MD,United States,2007:84-94 [11] 顾亦然,陈敏.一种三部图网络中标签时间加权的推荐方法[J].计算机科学,2012,9(8):96-98,9 [12] 黄创光,印鉴,汪静,等.不确定近邻的协同过滤推荐算法[J].计算机学报,2010,33(8):1369-1377 [13] 陈健,印鉴.基于影响集的协作过滤推荐算法[J].软件学报,2007,18(7):1685-1694 [14] Liu X,Datta A,Rzadca K,et al.Stereo trust:A Group based personalized trust model[C]∥Proceedings of the 18th ACM Conference on Information and Knowledge Management.Hong Kong,China,2009:7-16 [15] Jamali M,Ester M.Trust Walker:A random walk model forcombining trust-based and item-based recommendation[C]∥Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Paris,France,2009:397-406 [16] 楼俊钢,申情,沈张果.软件可靠性预测的核函数方法[J].计算机科学,2012,9(4):145-148 [17] Ma H,King I,Lyu M R.Effective missing data prediction for collaborative filtering[C]∥Proceedings of the 30th Annual International ACM SIGIR Conference.Amsterdam,The Netherlands,2007:39-46 |
No related articles found! |
|