融合隐性特征的群体推荐方法研究

doi:10.11896/j.issn.1002-137X.2017.03.048

摘要/Abstract

摘要： 作为目前最成功的主流推荐方法,奇异值分解算法(SVD)将已知的海量数据建模并通过矩阵分解降维处理来得到有效信息；非负矩阵分解(NMF)则通过分解出非负矩阵元素来解释特征意义。这两种较为成功的方法均通过对显性反馈信息进行基于矩阵分解的处理得到用户的喜好信息来进行群体推荐。然而,仅凭用户的显性反馈信息有时无法准确反映用户的真实喜好。为解决上述问题,提出了一种针对这两种模型的改进方法,将隐性特征和基于隐性特征的群体权重计算方法融合进经典的矩阵分解算法,其中隐性特征可以完善用户的喜好信息,基于隐性特征的群体权重计算方法则根据群体的特点给予用户相应的权重,使得推荐的准确率得到提升。对该方法在KDD Cup 2012 Track1中的腾讯微博数据集上进行测试,实验结果表明在该数据集上融合方法的平均绝对偏差(MAE)和准确率要优于SVD算法与NMF算法,推荐的性能有较明显的提升。

关键词: 群体推荐,隐性特征,群体权重,平均绝对偏差

Abstract: As the most successful mainstream recommendation method,singular value decomposition (SVD) algorithm builds the model from known huge data and uses the matrix decomposition dimension reduction to get effective information,and non negative matrix factorization (NMF) uses the decomposition of nonnegative matrix elements to explain the meanings of characteristics.These two kinds of successful methods are based on matrix decomposition of explicit feedback information,and obtain the user’s preference information.However,they cannot accurately reflect the true preferences of the users only according to user’s explicit feedback.To solve the problem,this paper put forward an improved method for the two models,integrating the hidden features and weight calculation method based on hidden features into the classical matrix decomposition algorithm,the hidden features can perfect the information of user’s prefe-rences.Weight calculation method based on hidden features can judge the group characteristics and give the appropriate weight to users,which improve the recommendation accuracy.The method was tested on the Tencent micro blog data set in KDD Cup 2012 Track1.The results show that from the experimental standard of the MAE and the precision,the fusion method is better than SVD and NMF on this data set,and significantly improves the recommendation perfor-mance.

Key words: Group recommendation,Hidden features,Weights of the group,MAE

刘毅,钟忺,李琳. 融合隐性特征的群体推荐方法研究[J]. 计算机科学, 2017, 44(3): 231-236. https://doi.org/10.11896/j.issn.1002-137X.2017.03.048

LIU Yi, ZHONG Xian and LI Lin. Research on Method of Group Recommendation for Fusion of Hidden Features[J]. Computer Science, 2017, 44(3): 231-236. https://doi.org/10.11896/j.issn.1002-137X.2017.03.048

参考文献

[1] LIU Q,CHEN E H,XIONG H,et al.Enhancing collaborativefiltering by user interest expansion[J].IEEE Transactions on Systems,Man,and Cybernetics-part B:Cybernetics,2012,42(1):218-233.
[2] KONSTAN J A,RIEDL J T.Recommender systems:from algorithms to user experience[J].User Modeling and User-Adapted Interaction,2012,22(1):101-123.
[3] CACHEDA F,CARNEIRO V,FERNANDEZ D,et al.Comparison of collaborative filtering algorithms:limitations of current techniques and proposals for scalable,high-performance recommender systems[J].ACM Transactions on the Web,2011,5(2):1-33.
[4] ARMSTRONG J R,WADE J.Development of Systems Engineering Expertise[J].Procedia Computer Science,2015,44(2):689-698.
[5] ZHENG X,CHEN C C,HUNG J L,et al.A Hybrid Trust-based Recommender System for Online Communities of Practice[J].IEEE Transactions on Learning Technologies,2015,8(4):345-356.
[6] COLOMBO-MENDOZA L O,VALENCIA-GARCA R,RODR-GUEZ-GONZLEZ A,et al.RecomMetz:A context-aware knowledge-based mobile recommender system for movie showtimes[J].Expert Systems with Applications,2015,42(3):1202-1222.
[7] ZHU Y X,LV L Y.A review of the evaluation index of the re-commendation system[J].Journal of the University of Electro-nic Science and technology of University of science and techno-logy,2012,41(2):163-175.(in Chinese) 朱郁筱,吕林媛.推荐系统评价指标综述[J].电子科技大学学报,2012,41(2):163-175.
[8] TEJEDA-LORENTE ,PORCEL C,P EIS E,et al.A quality based recommender system to disseminate information in a university digital library[J].Information Sciences,2014,261(5):52-69.
[9] ANBARASU V,LINDA X,MAHALAKSHMI S.An EfficientRecommender System based on Collaborative Filtering[J].AsianJournal of Applied Sciences,2015,3(1).
[10] MIU P.Research on Weibo user interest model based on information push technology[D].Wuhan:Wuhan University of Technology,2012.(in Chinese) 谬平.基于微博用户兴趣模型的信息推送技术的研宄[D].武汉:武汉理工大学,2012.
[11] MONTUSCHI P,LAMBERTI F, GATTESCHI V,et al.A Semantic Recommender System for Adaptive Learning[J].It Professional,2015,17(5):50-58.
[12] MEHDI M,BOUGUILA N,BENTAHAR J.Probabilistic ap-proach for QoS-aware recommender system for trustworthy web service selection[J].Applied Intelligence,2014,1(2):503-524.
[13] YU Z W,LI L,WONG H S,et al.Probabilistic cluster structure ensemble[J].Information Sciences,2014,267(5):16-34.
[14] SIHME A,LEHOCINEM B,MINIAIH A.Batch Adsorption of Phenol From Industrial Waste Water Using Cereal By-Produ-cts As A New Adsrbent[J].Energy Procedia,2012,18(4):1135-1144.
[15] CAO Y M.Research and implementation of personalized news recommendation algorithm based on collaborative filtering [D].Beijing:Beijing University of Posts and Telecommunications 2013.(in Chinese) 曹一鸣.基于协同过滤的个性化新闻推荐算法的研究与实现[D].北京:北京邮电大学 2013.
[16] ALPHY A,PRABAKRAN S.A Dynamic Recommender System for Improved Web Usage Mining and CRM Using Swarm Intelligence[J].Scientific World Journal,2015,2015:1-16.
[17] XIA P Y.Collaborative filtering algorithm in personalized re-commendation technology[D].Qingdao:Ocean University of China,2011.(in Chinese) 夏培勇.个性化推荐技术中的协同过滤算法研究[D].青岛:中国海洋大学,2011.
[18] RAHMAN R M,SIDDIQUEE M R,HAIDER N.Movie Recommendation System Based on Fuzzy Inference System and Adaptive Neuro Fuzzy Inference System[J].International Journal of Fuzzy System Applications,2015,4(4):31-69.
[19] CHEN L,CHEN G,WANG F.Recommender systems based on user reviews:the state of the art[J].User Modeling and User-Adapted Interaction,2015,25(2):99-154.
[20] WANG L C,MENG X W,ZHANG Y J.Chinese Journal of context aware recommendation system[J].Software,2012(1):1-20.(in Chinese) 王立才,孟祥武,张玉洁.上下文感知推荐系统[J].软件学报,2012(1):1-20.
[21] HUANG Q.Personalized recommendation algorithm of network book resources[D].Chengdu:Southwest Jiaotong University,2014.(in Chinese) 黄琼,网络图书资源个性化推荐算法研究[D].成都:西南交通大学,2014.
[22] DING B Z.Collaborative filtering algorithm based on citation information[D].Jilin:Jilin University,2014.(in Chinese) 丁彬钊.基于引文信息的协同过滤算法研究[D].吉林:吉林大学,2014.
[23] CHEN N Y.Design and implementation of recommendation system based on personalized recommendation engine[D].Guangzhou:South China University of Technology,2012.(in Chinese) 陈诺言.基于个性化推荐引擎组合的推荐系统的设计与实现[D].广州:华南理工大学,2012.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed