该文提出了一种新颖的概率交易模型PTM,针对线下百货进行个性化的推荐。传统的推荐模型,如K-近邻算法、矩阵分解等,或者仅利用局部的数据,使得模型面临线下数据极大的稀疏性挑战,或者忽略百货数据中的交易维度,使得模型损失了同一交易中多商品共现的强相关信息,最终导致它们在面对线下百货推荐问题时性能低下。针对以上的问题,本模型从交易的维度出发,建模交易记录中的共现模式,并利用全局的交易数据来学习商品的相关分量,在此基础上推断出用户的兴趣分布,实现个性化的推荐。在真实的线下百货交易数据上的实验结果表明,该模型能够极大地提高线下百货领域个性化推荐的准确性。
Abstract
In this paper, we propose a novel probabilistic transaction model (PTM) for brand recommendation in the traditional shopping mall. Some existing algorithms, such as KNN based recommendation, take only local information into consideration and suffer from the sparse problem in offline transaction data. Some algorithms, such as matrix factorization based recommendation, take all transactions for each user as a whole and fail to discriminatethe co-concurrence between inter- and intra-transactions. To address these two issues, the PTM is designed to learn the latent representation of brands and transactions from all the brand co-occurrences in each transaction, and then the latent representation for each user could be derived for personalized recommendation. Experiment on real transaction data sets shows that PTM based recommendation outperforms the baselines.
关键词
PTM /
概率交易模型 /
品牌共现
{{custom_keyword}} /
Key words
PTM /
probabilistic transaction model /
co-concurrence
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Igor V Cades,Padhraic Smyth,Heikki Mannila.Probabilistic modeling of transaction data with applications to profiling visualization and prediction [C]// Proceedings of the 7th ACM SIGKDD international conference on knowledge discovery and data mining,2001:37-46.
[2] Michael J Pazzani, Daniel Billsus:Content-Based Recommendation Systems. The Adaptive Web[M],2007:325-341.
[3] Alexandrin Popescul, LyleH Ungar, David M Pennock, et al. Probabilistic Models for Unified Collaborative and Content-Based Recommendation in Sparse-Data Environments[C]//Proceedings of the 17th Conference on Uncertainty in Artificial Intelligence,2001:437-444.
[4] Marco de Gemmis Pasquale Lops Giovanni Semeraro PierpaoloBasile.Integrating tags in a semantic content-based recommender[C]// Proceedings of the 2008 ACM conference on Recommender systems, 2008: 163-170.
[5] Gediminas Adomavicius, Alexander Tuzhilin.Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions [J]. Journal IEEE Transactions on Knowledge and Data Engineering,2005:734-749.
[6] Rong Pan,Peter Dolog,Guandong Xu.KNN-Based Clustering for Improving Social Recommender Systems[J]. Lecture Notes in Computer Science, 2013: 115-125.
[7] Hao Ma,Haixuan Yang,Michael R.lyu,IrwinKing.SoRec: SoRec: social recommendation using probabilistic matrix factorization[C]// Proceedings of the 17th ACM conference on information and knowledge management ,2008: 931-940.
[8] Steffen Renlde, Christoph Freudenthaler, Zeno Gantner,et al. Bayesian Personalized Ranking from Implicit Feedback[C]// Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence,2009:452-461.
[9] Chong Wang,David M Blei.Collaborative topic modeling for recommending scientific articles[C]// Proceedings of the 17th ACM SIGKDD international conference on knowledge discovery and data mining ,2011:448-456.
[10] Xiaohuia Yan,Jiafeng Guo,Yanyan Lan,et al. A Biterm Topic Model for Short Texts[C]// Proceedings of the 22nd international conference on World Wide Web,2013: 1445-1456.
[11] Osama Khalifa, David W Corne, Mike J Chantler, Fraser Halley: Multi-objective Topic Modeling[C]// Evolutionary Multi-Criterion Optimization,2013: 51-65.
[12] Steffen Rendle,Christoph Freudenthaler,Lars Schmidt-Thieme.Factorizing personalized Markov chains for next-basket recommendation[C]// Proceedings of the 19th International Conference on World Wide Web,2010:811-820.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
973课题(2012CB316303,2014CB340401);863课题(2014AA015204,2012AA011003)
{{custom_fund}}
PDF(1479 KB)
