计算机科学技术学报 ›› 2020,Vol. 35 ›› Issue (2): 281-294.doi: 10.1007/s11390-020-9956-9

• • 上一篇    下一篇

联合结构信息与时间信息的动态社交推荐方法

Yang Liu, Zhi Li, Wei Huang, Tong Xu*, En-Hong Chen, Fellow, CCF, Senior Member, IEEE   

  1. Anhui Province Key Laboratory of Big Data Analysis and Application, University of Science and Technology of China Hefei 230027, China
  • 收稿日期:2019-08-20 修回日期:2019-12-22 出版日期:2020-03-05 发布日期:2020-03-18
  • 通讯作者: Tong Xu E-mail:tongxu@ustc.edu.cn
  • 作者简介:Yang Liu received her B.E. degree in information security from University of Science and Technology of China (USTC), Hefei, in 2016. She is currently working toward her Master's degree in the Anhui Province Key Laboratory of Big Data Analysis and Application, USTC, Hefei. Her research interests include deep learning and its application in recommender systems.
  • 基金资助:
    This work was partially supported by the National Key Research and Development Program of China under Grant No. 2018YFB1402600, the National Natural Science Foundation of China under Grant Nos. 61703386 and U1605251, and the MSRA (Microsoft Research Asia) Collaborative Research Project.

Exploiting Structural and Temporal Influence for Dynamic Social-Aware Recommendation

Yang Liu, Zhi Li, Wei Huang, Tong Xu*, En-Hong Chen, Fellow, CCF, Senior Member, IEEE        

  1. Anhui Province Key Laboratory of Big Data Analysis and Application, University of Science and Technology of China Hefei 230027, China
  • Received:2019-08-20 Revised:2019-12-22 Online:2020-03-05 Published:2020-03-18
  • Contact: Tong Xu E-mail:tongxu@ustc.edu.cn
  • About author:Yang Liu received her B.E. degree in information security from University of Science and Technology of China (USTC), Hefei, in 2016. She is currently working toward her Master's degree in the Anhui Province Key Laboratory of Big Data Analysis and Application, USTC, Hefei. Her research interests include deep learning and its application in recommender systems.
  • Supported by:
    This work was partially supported by the National Key Research and Development Program of China under Grant No. 2018YFB1402600, the National Natural Science Foundation of China under Grant Nos. 61703386 and U1605251, and the MSRA (Microsoft Research Asia) Collaborative Research Project.

为更好的向用户提供智能的、个性化的推荐,已有大量研究在学习用户偏好表征时,尝试将社交背景信息融合进用户偏好建模过程,进而实现更精确的推荐。大多数已有的方法通常处理静态的社交信息,而在实际应用场景中,用户行为不断变化,社交关系也动态改变。为探究动态的社交信息对用户变化的消费偏好的影响,本文提出一种结合结构信息与时间信息的动态社交推荐方法。该方法由两部分组成:社交影响力提取模块(SIL)与用户个人偏好建模模块(DPL)。在SIL中,我们将社交关系组织成社交图序列,利用图卷积网络学习社交背景信息,并且,我们设计注意力机制从结构性与时间性两方面学习社交权重。在DPL模块中,我们利用循环神经网络建模用户个人消费偏好。最终的用户表征来自于两个模块的结合。我们将提出的方法与几种已有的经典算法进行对比实验,实验使用的两个数据集收集自现实社交平台(Epinions和Gowalla),使用的评价指标为命中率和归一化折损累积增益(HR@10和NDCG@10)。实验结果显示,提出的方法在Epinions数据集上HR@10和NDCG@10分别高于对照方法8.06%-44.65%和4.07%-34.81%,在Gowalla数据集上则提升分别为7.97%-35.86%和7.59%-31.21%。此外,本工作中对所设计的注意力机制中的结构注意力机制和时间注意力机制进行了分析,结果显示在社交背景信息中,同时考虑两类信息能够更有效地表征用户,进而提升推荐效果。

关键词: 推荐系统, 社交影响力, 序列推荐

Abstract: Recent years have witnessed the rapid development of online social platforms, which effectively support the business intelligence and provide services for massive users. Along this line, large efforts have been made on the socialaware recommendation task, i.e., leveraging social contextual information to improve recommendation performance. Most existing methods have treated social relations in a static way, but the dynamic influence of social contextual information on users' consumption choices has been largely unexploited. To that end, in this paper, we conduct a comprehensive study to reveal the dynamic social influence on users' preferences, and then we propose a deep model called Dynamic Social-Aware Recommender System (DSRS) to integrate the users' structural and temporal social contexts to address the dynamic socialaware recommendation task. DSRS consists of two main components, i.e., the social influence learning (SIL) and dynamic preference learning (DPL). Specifically, in the SIL module, we arrange social graphs in a sequential order and borrow the power of graph convolution networks (GCNs) to learn social context. Moreover, we design a structural-temporal attention mechanism to discriminatively model the structural social influence and the temporal social influence. Then, in the DPL part, users' individual preferences are learned dynamically by recurrent neural networks (RNNs). Finally, with a prediction layer, we combine the users' social context and dynamic preferences to generate recommendations. We conduct extensive experiments on two real-world datasets, and the experimental results demonstrate the superiority and effectiveness of our proposed model compared with the state-of-the-art methods.

Key words: recommender system, social influence, sequential recommendation

[1] Mnih A, Salakhutdinov R R. Probabilistic matrix factorization. In Proc. the 21st Annual Conference on Neural Information Processing Systems, December 2007, pp.1257-1264.
[2] Koren Y. Factorization meets the neighborhood:A multifaceted collaborative filtering model. In Proc. the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2008, pp.426-434.
[3] Rendle S, Freudenthaler C, Gantner Z et al. BPR:Bayesian personalized ranking from implicit feedback. In Proc. the 25th Conference on Uncertainty in Artificial Intelligence, June 2009, pp.452-461.
[4] Ma H, Yang H, Lyu M R et al. SoRec:Social recommendation using probabilistic matrix factorization. In Proc. the 17th ACM Conference on Information and Knowledge Management, October 2008, pp.931-940.
[5] Ma H, Zhou D, Liu C et al. Recommender systems with social regularization. In Proc. the 4th ACM International Conference on Web Search and Data Mining, February 2011, pp.287-296.
[6] Jamali M, Ester M. A matrix factorization technique with trust propagation for recommendation in social networks. In Proc. the 4th ACM Conference on Recommender Systems, September 2010, pp.135-142.
[7] Guo G, Zhang J, Yorke-Smith N. TrustSVD:Collaborative filtering with both the explicit and implicit influence of user trust and of item ratings. In Proc. the 29th AAAI Conference on Artificial Intelligence, January 2015, pp.123-129.
[8] Jiang M, Cui P, Wang F et al. Scalable recommendation with social contextual information. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(11):2789-2802.
[9] Biao C, Tong X, Qi L et al. Study on information diffusion analysis in social networks and its applications. International Journal of Automation and Computing, 2018, 15(4):377-401.
[10] Huang Z, Pan Z, Liu Q et al. An Ad CTR prediction method based on feature learning of deep and shallow layers. In Proc. the 2017 ACM on Conference on Information and Knowledge Management, November 2017, pp.2119-2122.
[11] Guo H, Tang R, Ye Y et al. DeepFM:A factorizationmachine based neural network for CTR prediction. arXiv:1703.04247, 2017. https://arxiv.org/abs/1703.04247,Dec.2019.
[12] He X, Liao L, Zhang H et al. Neural collaborative filtering. In Proc. the 26th International Conference on World Wide Web, April 2017, pp.173-182.
[13] Yu F, Liu Q, Wu S et al. A dynamic recurrent model for next basket recommendation. In Proc. the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval, July 2016, pp.729-732.
[14] Hidasi B, Karatzoglou A, Baltrunas L et al. Sessionbased recommendations with recurrent neural networks. arXiv:1511.06939, 2015. https://arxiv.org/abs/1511.06939,Dec.2019.
[15] Li J, Ren P, Chen Z et al. Neural attentive session-based recommendation. In Proc. the 2017 ACM Conference on Information and Knowledge Management, November 2017, pp.1419-1428.
[16] Hidasi B, Karatzoglou A. Recurrent neural networks with top-k gains for session-based recommendations. In Proc. the 27th ACM International Conference on Information and Knowledge Management, October 2018, pp.843-852.
[17] Lei C, Liu D, Li W et al. Comparative deep learning of hybrid representations for image recommendations. In Proc. the 2016 IEEE Conference on Computer Vision and Pattern Recognition, June 2016, pp.2545-2553.
[18] Hou M, Wu L, Chen E et al. Explainable fashion recommendation:A semantic attribute region guided approach. arXiv:1905.12862, 2019. https://arxiv.org/pdf/1905.12862.pdf,Dec.2019.
[19] Kim D, Park C, Oh J et al. Convolutional matrix factorization for document context-aware recommendation. In Proc. the 10th ACM Conference on Recommender Systems, September 2016, pp.233-240.
[20] Wang Q, Li S, Chen G. Word-driven and context-aware review modeling for recommendation. In Proc. the 27th ACM International Conference on Information and Knowledge Management, October 2018, pp.1859-1862.
[21] Aral S, Muchnik L, Sundararajan A. Distinguishing influence-based contagion from homophily-driven diffusion in dynamic networks. Proceedings of the National Academy of Sciences, 2009, 106(51):21544-21549.
[22] Sun P, Wu L, Wang M. Attentive recurrent social recommendation. In Proc. the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, July 2018, pp.185-194.
[23] Wu L, Sun P, Hong R et al. Collaborative neural social recommendation. IEEE Transactions on Systems, Man, and Cybernetics:Systems. doi:10.1109/TSMC.2018.2872842.
[24] Wu L, Sun P, Hong R et al. SocialGCN:An efficient graph convolutional network based model for social recommendation. arXiv:1811.02815, 2018. https://arxiv.org/abs/1811.02815,Dec.2019.
[25] Fan W, Ma Y, Li Q et al. Graph neural networks for social recommendation. arXiv:1902.07243, 2019. https://arxiv.org/pdf/1902.07243.pdf,Dec.2019.
[26] Qiu J, Tang J, Ma H et al. DeepInf:Social influence prediction with deep learning. In Proc. the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2018, pp.2110-2119.
[27] Wang H, Xu T, Liu Q et al. MCNE:An end-to-end framework for learning multiple conditional network representations of social network. In Proc. the 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, August 2019, pp.1064-1072.
[28] Xu T, Zhu H S, Zhong H et al. Exploiting the dynamic mutual influence for predicting social event participation. IEEE Transactions on Knowledge and Data Engineering, 2019, 31(6):1122-1135.
[29] Rendle S, Freudenthaler C, Schmidt-Thieme L. Factorizing personalized Markov chains for next-basket recommendation. In Proc. the 19th International Conference on World Wide Web, April 2010, pp.811-820.
[30] Wang P, Guo J, Lan Y et al. Learning hierarchical representation model for next basket recommendation. In Proc. the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, August 2015, pp.403-412.
[31] Wu C Y, Ahmed A, Beutel A et al. Recurrent recommender networks. In Proc. the 10th ACM International Conference on Web Search and Data Mining, February 2017, pp.495-503.
[32] Li Z, Zhao H, Liu Q et al. Learning from history and present:Next-item recommendation via discriminatively exploiting user behaviors. In Proc. the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2018, pp.1734-1743.
[33] Sordoni A, Bengio Y, Vahabi H et al. A hierarchical recurrent encoder-decoder for generative context-aware query suggestion. In Proc. the 24th ACM International on Conference on Information and Knowledge Management, October 2015, pp.553-562.
[34] Manotumruksa J, Macdonald C, Ounis I. A contextual attention recurrent architecture for context-aware venue recommendation. In Proc. the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, July 2018, pp.555-564.
[35] Smirnova E, Vasile F. Contextual sequence modeling for recommendation with recurrent neural networks. In Proc. the 2nd Workshop on Deep Learning for Recommender Systems, August 2017, pp.2-9.
[36] Manotumruksa J, Macdonald C, Ounis I. A contextual attention recurrent architecture for context-aware venue recommendation. In Proc. the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, July 2018, pp.555-564.
[37] Bruna J, Zaremba W, Szlam A et al. Spectral networks and locally connected networks on graphs. arXiv:1312.6203, 2013. https://arxiv.org/abs/1312.6203,Dec.2013.
[38] Feng C, Liu Z, Lin S et al. Attention-based graph convolutional network for recommendation system. In Proc. the 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, May 2019, pp.7560-7564.
[39] Zhang J, Shi X, Xie J et al. GaAN:Gated attention networks for learning on large and spatiotemporal graphs. arXiv:1803.07294, 2018. https://arxiv.org/pdf/1803.07294.pdf,Dec.2019.
[40] Veličković P, Cucurull G, Casanova A et al. Graph attention networks. arXiv:1710.10903, 2017. https://arxiv.org/abs/1710.10903,Dec.2019.
[41] Wu L K, Li Z, Zhao H K et al. Estimating early fundraising performance of innovations via graph-based market environment model. arXiv:1912.06767, 2019. https://arxiv.org/abs/1912.06767v1,Dec.2019.
[42] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Computation, 1997, 9(8):1735-1780.
[43] Kingma D P, Ba J. Adam:A method for stochastic optimization. arXiv:1412.6980, 2014. https://arxiv.org/abs/1412.6980,Dec.2019.
[44] Richardson M, Agrawal R, Domingos P. Trust management for the semantic Web. In Proc. the 2nd International Semantic Web Conference, October 2003, pp.351-368.
[45] Scellato S, Noulas A, Mascolo C. Exploiting place features in link prediction on location-based social networks. In Proc. the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2011, pp.1046-1054.
[1] Reza Jafari Ziarani, Reza Ravanmehr. 推荐系统中的意外效应:系统文献综述[J]. 计算机科学技术学报, 2021, 36(2): 375-396.
[2] Qi Liu, Hong-Ke Zhao, Le Wu, Zhi Li, En-Hong Chen. 通过物品的显式关系理解推荐系统[J]. , 2018, 33(4): 739-755.
[3] Ming-Xin Gan, Lily Sun, Rui Jiang. Trinity——一种基于用户-对象-标签异质网络游走的个性化推荐方法[J]. , 2016, 31(3): 577-594.
[4] Xin Xin, Chin-Yew Lin, Xiao-Chi Wei, He-Yan Huang. 分解遇见异构潜在主题:一个可解释的跨网站推荐模型[J]. , 2015, 30(4): 917-932.
[5] Xiang-Liang Zhang, Tak Man Desmond Lee, and Georgios Pitsilis. 防护推荐系统免遭托攻击[J]. , 2013, 28(4): 616-624.
[6] Hui-Feng Sun1 (孙慧峰), Student Member, CCF, ACM, Jun-Liang Chen1 (陈俊亮), Gang Yu1 (俞钢), Member, IEEE, Chuan-Chang Liu1 (刘传昌), Member, IEEE, Yong Peng1 (彭泳), Guang Chen2 (陈光), and Bo Cheng1 (程渤), Senior Member, CCF, Member, ACM. JacUOD:协同过滤中一种新的相似性度量方法[J]. , 2012, 27(6): 1252-1260.
[7] Jun Hu (胡军), Bing Wang (王兵), Yu Liu (刘禹), Member, CCF, and De-Yi Li (李德毅), Fello. 使用社会影响力进行个性化标签推荐[J]. , 2012, 27(3): 527-540.
[8] Marcelo G. Armentano, Daniela Godoy, and Analia Amandi. 微博社区中基于拓扑的用户推荐[J]. , 2012, 27(3): 624-634.
[9] Kiatichai Treerattanapitak and Chuleerat Jaruskulchai. 基于指数模糊C均值的协同过滤[J]. , 2012, 27(3): 567-576.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 周笛;. A Recovery Technique for Distributed Communicating Process Systems[J]. , 1986, 1(2): 34 -43 .
[2] 陈世华;. On the Structure of Finite Automata of Which M Is an(Weak)Inverse with Delay τ[J]. , 1986, 1(2): 54 -59 .
[3] 李万学;. Almost Optimal Dynamic 2-3 Trees[J]. , 1986, 1(2): 60 -71 .
[4] C.Y.Chung; 华宣仁;. A Chinese Information Processing System[J]. , 1986, 1(2): 15 -24 .
[5] 章萃; 赵沁平; 徐家福;. Kernel Language KLND[J]. , 1986, 1(3): 65 -79 .
[6] 王建潮; 魏道政;. An Effective Test Generation Algorithm for Combinational Circuits[J]. , 1986, 1(4): 1 -16 .
[7] 陈肇雄; 高庆狮;. A Substitution Based Model for the Implementation of PROLOG——The Design and Implementation of LPROLOG[J]. , 1986, 1(4): 17 -26 .
[8] 黄河燕;. A Parallel Implementation Model of HPARLOG[J]. , 1986, 1(4): 27 -38 .
[9] 郑国梁; 李辉;. The Design and Implementation of the Syntax-Directed Editor Generator(SEG)[J]. , 1986, 1(4): 39 -48 .
[10] 黄学东; 蔡莲红; 方棣棠; 迟边进; 周立; 蒋力;. A Computer System for Chinese Character Speech Input[J]. , 1986, 1(4): 75 -83 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn
总访问量: