Skip to main content
SpringerLink
Log in

From edge data to recommendation: A double attention-based deformable convolutional network

  • Published:
Peer-to-Peer Networking and Applications Aims and scope Submit manuscript

Abstract

Recommender systems (RSs) have become crucial parts in most web-scale applications, and the data sparsity is still one of the serious problems in RSs. Recently, the data sparsity can be tremendously alleviated by making use of informative data obtained from applications of edge devices and deep learning technologies with powerful data processing ability. In the face of extremely sparse rating data, the rich semantic information in reviews and the powerful feature extraction ability of convolutional neural network (CNN) contribute greatly to the improvement of recommendation performance. However, due to the complexity of natural language semantic components, it is common for many phrases in reviews to be separated by other words. Therefore, for the semantic information with unfixed intervals, the fixed geometric structure of CNN may lead to insufficient understanding of the user intention. Moreover, the usefulness of different reviews and the importance of different words in each review are various, which is vital for accurate modeling. In this paper, we propose a Double Attention-based Deformable Convolutional Network called DADCN for recommendation. In the proposed DADCN, two parallel deformable convolutional networks, which adopt the word-level and review-level attention mechanisms, are designed to flexibly extract features of both users and items from reviews. The parallel deformable convolutional networks jointly learn user preferences and item attributes, which is helpful to deepen the understanding of users’ attitudes. The word-level and review-level attention mechanisms are applied to intensify the critical words and informative reviews by assigning relatively high attention weights to them. Extensive experimental results on four real-world datasets demonstrate that the proposed DADCN outperforms four baseline methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  1. Ai X, Chen H, Lin K, Wang Z, Yu J (2020) Nowhere to Hide: Efficiently Identifying Probabilistic Cloning Attacks in Large-Scale RFID Systems, IEEE Transactions on Information Forensics and Security, DOI: https://doi.org/10.1109/TIFS.2020.3023785

  2. Chen H, Wang Z, Xia F, Li Y, Shi L (2018) Efficiently and completely identifying missing key tags for anonymous RFID systems. IEEE Internet Things J 5(4):2915–2926

    Article  Google Scholar 

  3. Chen H, Xue G, Wang Z (2017) Efficient and reliable missing tag identification for large-scale RFID systems with unknown tags. IEEE Internet Things J 4(3):736–748

    Article  Google Scholar 

  4. Imanbayev K, Sinchev B, Sibanbayeva S, Mukhanova A, Nurgulzhanov A, Zaurbekov N, Zaurbekova N, Korolyova NV, Baibolova L (2020) Analysis and mathematical modeling of big data processing,” Peer-to-Peer Networking and Applications

  5. Warren J (2015) Big data: principles and best practices of scalable Realtime data systems, Manning

  6. Altulyan M, Yao L, Wang X, Huang C, Kanhere SS, Sheng QZ (2020) Recommender Systems for the Internet of Things: A Survey, arXiv

  7. Kong X, Xia F, Wang J, Rahim A, Das SK (2017) Time-location-relationship combined service recommendation based on taxi trajectory data. IEEE Trans Ind Inf 13(3):1202–1212

    Article  Google Scholar 

  8. Xia F, Liu H, Lee I, Cao L (2016) Scientific article recommendation: exploiting common author relations and historical preferences. IEEE Trans Big Data 2(2):101–112

    Article  Google Scholar 

  9. Xia F, Liu J, Nie H, Fu Y, Wan L, Kong X (2020) Random walks: a review of algorithms and applications. IEEE Trans Emerg Topics Comput Intell 4(2):95–107

    Article  Google Scholar 

  10. Gong Y, Jiang Z, Feng Y, Hu B, Zhao K, Liu Q, Ou W (2020) EdgeRec: recommender system on edge in Mobile Taobao, in Proc. of ACM CIKM

  11. Koren Y (2008) Factorization meets the neighborhood: a multifaceted collaborative filtering model, in Proc. of ACM SIGKDD

  12. Koren Y, Bell R, Volinsky C (2009) Matrix factorization techniques for recommender systems. Computer 42:30–37

    Article  Google Scholar 

  13. Salakhutdinov R, Mnih A (2008) Probabilistic matrix factorization, Adv Neural Inf Proces Syst, 1257–1264

  14. Lee DD, Seung HS (2001) Algorithms for non-negative matrix factorization, Adv Neural Inf Proces Syst, 556–562

  15. Zhang F, Yuan N, Ang DL, Xie X, Ma W (2016) Collaborative Knowledge Base Embedding for Recommender Systems, in Proc. of ACM SIGKDD, 353–362

  16. McAuley J, Leskovec J (2013) Hidden Factors and Hidden Topics: Understanding Rating Dimensions with Review Text, in Proc. of ACM RecSys, 165–172

  17. Kalchbrenner N, Grefenstette E, Blunsom P (2014) A convolutional neural network for modelling sentences, in Proc. of ACL

  18. Dai J, Qi H, Xiong Y, L Yi, Zhang G, Hu H, Wei Y (2017) Deformable Convolutional Networks, in Proc. of IEEE ICCV, 764–773

  19. Zheng L, Noroozi V, Yu P (2017) Joint deep modeling of users and items using reviews for recommendation, in Proc. of ACM WSDM, 425–434

  20. Chen J, Zhang H, He X, Nie L, Liu W, Chua T (2017) Attentive collaborative fltering: Multimedia recommendation with item- and component-level attention, in Proc. of ACM SIGIR, 335–344

  21. Chen C, Zhang M, Liu Y, Ma S (2018) Neural attentional rating regression with review-level explanations, in Proc. of WWW

  22. Liu H, Wu F, Wang W, Wang X, Jiao P, Wu C, Xie X (2019) NRPA: neural recommendation with personalized attention, in Proc. of ACM SIGIR

  23. H. Bharadhwaj, H. Park, and B. Lim (2018) RecGAN: recurrent generative adversarial networks for recommendation systems, in Proc. of ACM RecSys, 372–376

  24. Elkahky A, Song Y, He X (2015) A multi-view deep learning approach for cross domain user modeling in recommendation systems in Proc. of WWW, 278–288

  25. Hidasi B, Karatzoglou A, Baltrunas L, Tikk D (2016) Session-based recommendations with recurrent neural networks, in Proc. of ICLR

  26. Kim D, Park C, Oh J, Lee S, Yu H (2016) Convolutional Matrix Factorization for Document Context-Aware Recommendation, in Proc. of ACM RecSys, 233–240

  27. H. Chen, J. Fu, L. Zhang, S. Wang, and L. Wang (2019) Deformable convolutional matrix factorization for document context-aware recommendation in social networks, IEEE Access

  28. Pang G, Wang X, Hao F, Xie J, Qin X (2019) ACNN-FM: a novel recommender with attention-based convolutional neural network and factorization machines, Knowledge-Based System

  29. He X, Liao L, Zhang H, Nie L, Hu X, Chua T (2017) Neural collaborative fltering, in Proc. of WWW, 173–182

  30. Nair V, Hinton GE (2010) Rectifed Linear Units Improve Restricted Boltzmann Machines, in Proc. of ACM ICML, 807–814

  31. Li P, Wang Z, Ren Z, Bing L, Lam W (2017) Neural Rating Regression with Abstractive Tips Generation for Recommendation, in Proc. of ACM SIGIR, 345–354

  32. Kingma D, Ba J (2014) Adam: a method for stochastic Optimizatiion, Computer Science

  33. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting, J Mach Learn Res 1929–1958

Download references

Acknowledgements

This work was supported in part by NSFC grant (No.61772551), the International Exchange Project of China University of Petroleum (No.UPCIEF202052), the Major Scientific and Technological Projects of CNPC under Grant ZD2019-183-003.

Author information

Authors and Affiliations

  1. College of Control Science and Engineering, China University of Petroleum, Room 1215, Engineering Building E, No. 66, Changjiang West Road, Huangdao District, Qingdao, Shandong Province, People’s Republic of China

    Zhe Li, Honglong Chen & Kai Lin

  2. Novosibirsk State Technical University, Novosibirsk, Russia

    Vladimir Shakhov

  3. Collge of Computer Science and Technology, China University of Petroleum, Qingdao, People’s Republic of China

    Leyi Shi

  4. Shandong Computer Science Center (National Supercomputer Center in Jinan), Shandong Key Laboratory of Computer Networks, Qilu University of Technology (Shandong Academy of Sciences), Jinan, People’s Republic of China

    Jiguo Yu

Authors
  1. Zhe Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Honglong Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kai Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Vladimir Shakhov
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Leyi Shi
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jiguo Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Honglong Chen.

Additional information

This article is part of the Topical Collection: Special Issue on Convergence of Edge Computing and Next Generation Networking

Guest Editors: Deze Zeng, Geyong Min, Qiang He, and Song Guo

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, Z., Chen, H., Lin, K. et al. From edge data to recommendation: A double attention-based deformable convolutional network. Peer-to-Peer Netw. Appl. 14, 3984–3997 (2021). https://doi.org/10.1007/s12083-020-01037-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12083-020-01037-7

Keywords

Search

Navigation