计算机科学技术学报 ›› 2020,Vol. 35 ›› Issue (4): 794-808.doi: 10.1007/s11390-020-0314-8

所属专题： Data Management and Data Mining

• • 上一篇    下一篇

ATLRec:用于跨领域推荐的注意力对抗迁移学习网络

Ying Li, Member, CCF, Jia-Jie Xu*, Member, CCF, ACM, Peng-Peng Zhao, Member, CCF, ACM, IEEE Jun-Hua Fang, Wei Chen, Member, CCF, Lei Zhao, Member, CCF, ACM   

  1. School of Computer Science and Technology, Soochow University, Suzhou 215006, China
  • 收稿日期:2020-01-18 修回日期:2020-06-03 出版日期:2020-07-20 发布日期:2020-07-20
  • 通讯作者: Jia-Jie Xu E-mail:xujj@suda.edu.cn
  • 作者简介:Ying Li received her B.S. degree in computer science from Soochow University, Suzhou, in 2018. She is currently a Master student in the School of Computer Science and Technology at Soochow University, Suzhou. Her research interests include data mining and recommender systems.
  • 基金资助:
    This work was supported by the National Natural Science Foundation of China under Grant Nos. 61872258, 61772356, 61876117, and 61802273, and the Priority Academic Program Development of Jiangsu Higher Education Institutions of China.

ATLRec: An Attentional Adversarial Transfer Learning Network for Cross-Domain Recommendation

Ying Li, Member, CCF, Jia-Jie Xu*, Member, CCF, ACM, Peng-Peng Zhao, Member, CCF, ACM, IEEE Jun-Hua Fang, Wei Chen, Member, CCF, Lei Zhao, Member, CCF, ACM        

  1. School of Computer Science and Technology, Soochow University, Suzhou 215006, China
  • Received:2020-01-18 Revised:2020-06-03 Online:2020-07-20 Published:2020-07-20
  • Contact: Jia-Jie Xu E-mail:xujj@suda.edu.cn
  • About author:Ying Li received her B.S. degree in computer science from Soochow University, Suzhou, in 2018. She is currently a Master student in the School of Computer Science and Technology at Soochow University, Suzhou. Her research interests include data mining and recommender systems.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China under Grant Nos. 61872258, 61772356, 61876117, and 61802273, and the Priority Academic Program Development of Jiangsu Higher Education Institutions of China.

PDF (PC)

57

补充材料

1

考虑到单域推荐中通常遇到的数据稀疏性和冷启动问题，通过实体链接将不同的域链接在一起，为更有效、更准确地实现推荐任务和其他任务提供了新的机会。利用来自辅助域的知识，可以提高目标域内的推荐性能。虽然已经有工作提出了一些跨领域的推荐方法来提高目标领域的推荐性能，但是其中大多数现有的基于迁移的方法更关注如何转移，而不是如何获得领域共享的特性，这导致了特定于某一单一领域的特性被转移从而造成次优结果。虽然有些方法考虑了领域特定的特性，但大多数都是浅层模型，无法学习复杂的非线性用户-项目交互关系。现有的一些基于内容以及基于嵌入的方法也更依赖于辅助内容，如文本数据和可视数据。我们所提出的方法ATLRec受对抗转移学习的启发，仅使用来自两个域的隐式反馈的信息，就可以有效地捕获要转移的域共享特性和每个域的特定于域的特性。在ATLRec中，我们利用对抗性学习来生成两个域的用户-项交互的表示，从而使鉴别器不能识别它们属于哪个域，以获得域共享的特性。同时，每个域通过一个私有特征提取器学习其特定于该领域的特征。每个域的推荐都考虑了领域特有特征以及领域共享特征。我们进一步采用注意力机制，利用具有交互历史的共享用户来学习两个域的项的潜在因素，以便在共享空间中充分学习所有项的表示，即使不同域共享的项很少甚至没有。通过这种方法，我们可以在共享空间中表示来自源域和目标域的所有项，以便更好地链接不同域中的项，并捕获跨域项与项之间的相关性，从而促进域共享知识的学习。我们在各种真实数据集上进行了实验，实验结果证明该模型在推荐准确度等方面优于最先进的单域和跨域推荐方法。

关键词: 对抗迁移学习, 注意力机制, 跨领域推荐, 实体链接

Abstract: Entity linking is a new technique in recommender systems to link users' interaction behaviors in different domains, for the purpose of improving the performance of the recommendation task. Linking-based cross-domain recommendation aims to alleviate the data sparse problem by utilizing the domain-sharable knowledge from auxiliary domains. However, existing methods fail to prevent domain-specific features to be transferred, resulting in suboptimal results. In this paper, we aim to address this issue by proposing an adversarial transfer learning based model ATLRec, which effectively captures domain-sharable features for cross-domain recommendation. In ATLRec, we leverage adversarial learning to generate representations of user-item interactions in both the source and the target domains, such that the discriminator cannot identify which domain they belong to, for the purpose of obtaining domain-sharable features. Meanwhile each domain learns its domain-specific features by a private feature extractor. The recommendation of each domain considers both domain-specific and domain-sharable features. We further adopt an attention mechanism to learn item latent factors of both domains by utilizing the shared users with interaction history, so that the representations of all items can be learned sufficiently in a shared space, even when few or even no items are shared by different domains. By this method, we can represent all items from the source and the target domains in a shared space, for the purpose of better linking items in different domains and capturing cross-domain item-item relatedness to facilitate the learning of domain-sharable knowledge. The proposed model is evaluated on various real-world datasets and demonstrated to outperform several state-of-the-art single-domain and cross-domain recommendation methods in terms of recommendation accuracy.

Key words: adversarial transfer learning, attention mechanism, cross-domain recommendation, entity linking

[1] Srivastava R, Palshikar G K, Chaurasia S, Dixit A M. What's next? A recommendation system for industrial training. Data Science and Engineering, 2018, 3(3):232-247.
[2] Zhao P, Zhu H, Liu Y, Xu J, Li Z, Zhuang F, Sheng V S, Zhou X. Where to go next:A spatio-temporal gated network for next POI recommendation. In Proc. the 33rd AAAI Int. Conf. Artificial Intelligence, January 2019, pp.5877-5884.
[3] Li B, Yang Q, Xue X. Can movies and books collaborate? Cross-domain collaborative filtering for sparsity reduction. In Proc. the 21st Int. Joint Conf. Artificial Intelligence, July 2009, pp.2052-2057.
[4] Cantador I, Fernández-Tobías I, Berkovsky S, Cremonesi P. Cross-domain recommender systems. In Recommender Systems Handbook, Ricci F, Rokach L, Shapira B (eds.), Springer, 2015, pp.919-959.
[5] Liu J, Zhao P, Zhuang F, Liu Y, Sheng V S, Xu J, Zhou X, Xiong H. Exploiting aesthetic preference in deep cross networks for cross-domain recommendation. In Proc. the 29th World Wide Web Conference, April 2020, pp.2768-2774.
[6] Berkovsky S, Kuik T, Ricci F. Mediation of user models for enhanced personalization in recommender systems. User Model. User-Adapt. Interact., 2008, 18(3):245-286.
[7] Fernández-Tobías I, Cantador I. Exploiting social tags in matrix factorization models for cross-domain collaborative filtering. In Proc. the 1st Workshop on New Trends in Content-based Recommender Systems Co-Located with the 8th ACM Conf. Recommender Systems, October 2014, pp.34-41.
[8] Tan S, Bu J, Qin X, Chen C, Cai D. Cross domain recommendation based on multi-type media fusion. Neurocomputing, 2014, 127:124-134.
[9] Man T, Shen H, Jin X, Cheng X. Cross-domain recommendation:An embedding and mapping approach. In Proc. the 26th Int. Joint Conf. Artificial Intelligence, August 2017, pp.2464-2470.
[10] Zhu F, Wang Y, Chen C, Liu G, Orgun M A, Wu J. A deep framework for cross-domain and cross-system recommendations. In Proc. the 27th Int. Joint Conf. Artificial Intelligence, July 2018, pp.3711-3717.
[11] Yang Q, Hu G, Zhang Y. CoNet:Collaborative cross networks for cross-domain recommendation. In Proc. the 27th ACM Int. Conf. Information and Knowledge Management, October 2018, pp.667-676.
[12] Huang L, Zhao Z L, Wang C D, Huang D, Chao H Y. LSCD:Low-rank and sparse cross-domain recommendation. Neurocomputing, 2019, 366:86-96.
[13] Loni B, Shi Y, Larson M A, Hanjalic A. Cross-domain collaborative filtering with factorization machines. In Proc. the 36th European Conf. Information Retrieval Research, April 2014, pp.656-661.
[14] Zhang F, Yuan N J, Lian D, Xie X, Ma W Y. Collaborative knowledge base embedding for recommender systems. In Proc. the 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, August 2016, pp.353-362.
[15] Wang W, Yin H, Du X, Hua W, Li Y, Nguyen Q V H. Online user representation learning across heterogeneous social networks. In Proc. the 42nd ACM SIGIR Int. Conf. Research and Development in Information Retrieval, July 2019, pp.545-554.
[16] Ma J, Wen J, Zhong M, Chen W, Li X. MMM:Multi-source multi-net micro-video recommendation with clustered hidden item representation learning. Data Science and Engineering, 2019, 4(3):240-253.
[17] Tan C, Sun F, Kong T, Zhang W, Yang C, Liu C. A survey on deep transfer learning. In Proc. the 27th Int. Conf. Artificial Neural Networks, October 2018, pp.270-279.
[18] Goodfellow I J, Pouget-Abadie J, Mirza M, Xu B, WardeFarley D, Ozair S, Courville A C, Bengio Y. Generative adversarial nets. In Proc. the 2014 Annual Conf. Neural Information Processing Systems, December 2014, pp.2672-2680.
[19] Yuan F, Yao L, Benatallah B. DARec:Deep domain adaptation for cross-domain recommendation via transferring rating patterns. In Proc. the 28th Int. Joint Conf. Artificial Intelligence, August 2019, pp.4227-4233.
[20] Wang C, Niepert M, Li H. RecSys-DAN:Discriminative adversarial networks for cross-domain recommender systems. IEEE Transactions on Neural Networks and Learning Systems. doi:10.1109/TNNLS.2019.2907430.
[21] Kanagawa H, Kobayashi H, Shimizu N, Tagami Y, Suzuki T. Cross-domain recommendation via deep domain adaptation. In Proc. the 41st European Conf. Information Retrieval, April 2019, pp.20-29.
[22] He X, Liao L, Zhang H, Nie L, Hu X, Chua T S. Neural collaborative filtering. In Proc. the 26th Int. Conf. World Wide Web, April 2017, pp.173-182.
[23] Berkovsky S, Kuik T, Ricci F. Cross-domain mediation in collaborative filtering. In Proc. the 11th Int. Conf. User Modeling, June 2007, pp.355-359.
[24] Singh A P, Gordon G J. Relational learning via collective matrix factorization. In Proc. the 14th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, August 2008, pp.650-658.
[25] Denton E L, Chintala S, Szlam A, Fergus R. Deep generative image models using a Laplacian pyramid of adversarial networks. In Proc. the 2015 Annual Conf. Neural Information Processing Systems, December 2015, pp.1486-1494.
[26] Zhang Y, Barzilay R, Jaakkola T S. Aspect-augmented adversarial networks for domain adaptation. Transactions of the Association for Computational Linguistics, 2017, 5:515-528.
[27] Gui T, Zhang Q, Huang H, Peng M, Huang X. Part-ofspeech tagging for twitter with adversarial neural networks. In Proc. the 2017 Conf. Empirical Methods in Natural Language Processing, September 2017, pp.2411-2420.
[28] Xu C, Zhao P, Liu Y, Sheng V S, Xu J, Zhuang F, Fang J, Zhou X. Graph contextualized self-attention network for session-based recommendation. In Proc. the 28th Int. Joint Conf. Artificial Intelligence, August 2019, pp.3940-3946.
[29] Zhang T, Zhao P, Liu Y, Sheng V S, Xu J, Wang D, Liu G, Zhou X. Feature-level deeper self-attention network for sequential recommendation. In Proc. the 28th Int. Joint Conf. Artificial Intelligence, August 2019, pp.4320-4326.
[30] Chen J, Zhang H, He X, Nie L, Liu W, Chua T S. Attentive collaborative filtering:Multimedia recommendation with item- and component-level attention. In Proc. the 40th ACM SIGIR Int. Conf. Research and Development in Information Retrieval, August 2017, pp.335-344.
[31] Cheng Z, Ding Y, He X, Zhu L, Song X, Kankanhalli M S. A3NCF:An adaptive aspect attention model for rating prediction. In Proc. the 27th Int. Joint Conf. Artificial Intelligence, July 2018, pp.3748-3754.
[32] Shi S, Zhang M, Liu Y, Ma S. Attention-based adaptive model to unify warm and cold starts recommendation. In Proc. the 27th ACM Int. Conf. Information and Knowledge Management, October 2018, pp.127-136.
[33] Xi W D, Huang L, Wang C D, Zheng Y Y, Lai J. B-PAM:Recommendation based on BP neural network with attention mechanism. In Proc. the 28th Int. Joint Conf. Artificial Intelligence, August 2019, pp.3905-3911.
[34] Hu Y, Koren Y, Volinsky C. Collaborative filtering for implicit feedback datasets. In Proc. the 8th IEEE Int. Conf. Data Mining, December 2008, pp.263-272.
[35] Pan R, Zhou Y, Cao B, Liu N N, Lukose R M, Scholz M, Yang Q. One-class collaborative filtering. In Proc. the 8th IEEE Int. Conf. Data Mining, December 2008, pp.502-511.
[36] Ganin Y, Lempitsky V S. Unsupervised domain adaptation by backpropagation. In Proc. the 32nd Int. Conf. Machine Learning, July 2015, pp.1180-1189.
[37] He R, McAuley J J. Ups and downs:Modeling the visual evolution of fashion trends with one-class collaborative filtering. In Proc. the 25th Int. Conf. World Wide Web, April 2016, pp.507-517.
[38] Rendle S, Freudenthaler C, Gantner Z, Schmidt-Thieme L. BPR:Bayesian personalized ranking from implicit feedback. In Proc. the 25th Conf. Uncertainty in Artificial Intelligence, June 2009, pp.452-461.
[39] Misra I, Shrivastava A, Gupta A, Hebert M. Cross-stitch networks for multi-task learning. In Proc. the 2016 IEEE Conf. Computer Vision and Pattern Recognition, June 2016, pp.3994-4003.
[1] Jia-Ke Ge, Yan-Feng Chai, Yun-Peng Chai. WATuning：一种基于注意力机制的深度强化学习的工作负载感知调优系统[J]. 计算机科学技术学报, 2021, 36(4): 741-761.
[2] Chen-Chen Sun, De-Rong Shen. 面向深度实体匹配的混合层次网络[J]. 计算机科学技术学报, 2021, 36(4): 822-838.
[3] Sheng-Luan Hou, Xi-Kun Huang, Chao-Qun Fei, Shu-Han Zhang, Yang-Yang Li, Qi-Lin Sun, Chuan-Qing Wang. 基于深度学习的文本摘要研究综述[J]. 计算机科学技术学报, 2021, 36(3): 633-663.
[4] Yang Liu, Ruili He, Xiaoqian Lv, Wei Wang, Xin Sun, Shengping Zhang. 婴儿的年龄和性别容易被识别吗？[J]. 计算机科学技术学报, 2021, 36(3): 508-519.
[5] Yi-Ting Wang, Jie Shen, Zhi-Xu Li, Qiang Yang, An Liu, Peng-Peng Zhao, Jia-Jie Xu, Lei Zhao, Xun-Jie Yang. 基于搜索引擎丰富上下文信息的实体链接方法[J]. 计算机科学技术学报, 2020, 35(4): 724-738.
[6] Huan-Jing Yue, Sheng Shen, Jing-Yu Yang, Hao-Feng Hu, Yan-Fang Chen. 基于渐进式通道注意力网络的参考图引导超分辨率研究[J]. 计算机科学技术学报, 2020, 35(3): 551-563.
[7] Chun-Yang Ruan, Ye Wang, Jiangang Ma, Yanchun Zhang, Xin-Tian Chen. 基于元路径注意力机制的异构网络对抗式嵌入[J]. 计算机科学技术学报, 2019, 34(6): 1217-1229.
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] 李未;. A Structural Operational Semantics for an Edison Like Language(2)[J]. , 1986, 1(2): 42 -53 .
[3] 章萃; 赵沁平; 徐家福;. Kernel Language KLND[J]. , 1986, 1(3): 65 -79 .
[4] 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 .
[5] 乔香珍;. An Efficient Parallel Algorithm for FFT[J]. , 1987, 2(3): 174 -190 .
[6] 李明慧;. CAD System of Microprogrammed Digital Systems[J]. , 1987, 2(3): 226 -235 .
[7] S.T.Chanson; 梁路平; A.Kumar;. Throughput Models of CSMA Network with Stations Uniformly Distributed along the Bus[J]. , 1987, 2(4): 243 -264 .
[8] 戚余禄;. A Systolic Approach for an Improvement of a Finite Field Multiplier[J]. , 1987, 2(4): 303 -309 .
[9] 黎仁蔚;. Soundness and Completeness of Kung s Reasoning Procedure[J]. , 1988, 3(1): 7 -15 .
[10] 包丰;. On the Condition for FSM Being a Scrambler[J]. , 1988, 3(1): 70 -74 .
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：