Advances and Perspectives on Knowledge Transfer Based Cross-Domain Recom-mendation

doi:10.3778/j.issn.1673-9418.2006053

Abstract

Abstract:

Data sparseness and cold start problems caused by unbalanced data distribution restrict the further development of personalized recommendation systems. With the rise of transfer learning technology, cross-domain recommendation based on transfer learning provides possibility to solve such problems. This kind of algorithm can solve the recommendation task in the target domain by transferring appropriate auxiliary domain knowledge which is different but related to the target domain, and improve the performance of target recommendation task in the target domain. The unique advantage of deep learning in non-linear feature learning and representation has greatly improved the performance of deep cross-domain recommendation algorithms. A review of cross-domain recommendation algorithms for knowledge transfer in recent years is carried out. The state-of-the-art algorithms are divided into two categories: cross-domain recommendation and deep cross-domain recommendation. According to different knowledge transfer technologies, they are sorted and summarized separately. And then various algorithms are analyzed and compared in depth from different perspectives such as model interpretability, applicable scenarios, user characteristics, model evaluation, etc. Finally, this paper summarizes the existing problems and dificiencies of existing algorithms, explores passible solutions and forecasts the future development trend.

Key words: data sparsity, transfer learning, knowledge transfer, cross-domain recommendation

摘要：

由数据分布不均衡产生的数据稀疏和冷启动问题制约着个性化推荐系统进一步发展。随着迁移学习技术兴起，基于迁移学习的跨领域推荐为解决该类问题提供了可能。面向知识迁移的跨领域推荐算法通过迁移与目标领域不同但相关的辅助知识来解决目标领域中的推荐任务，提高目标域的推荐性能。而深度学习在非线性特征的学习和表示上的独特优势进一步提升了深度跨域推荐的算法性能。对近年来面向知识迁移的跨领域推荐算法展开综述，将当前主流算法分为传统跨领域推荐算法和深度跨领域推荐算法两大类，又按照应用的不同知识迁移技术对两大类跨域推荐算法分别梳理和阐述，从模型的可解释性、适用场景、用户特征的描述能力、模型评价等不同角度对当前各类跨域推荐算法做出深度分析和比较，总结其存在的问题和不足，探讨可能的解决方法，展望未来的研究趋势。

关键词: 数据稀疏, 迁移学习, 知识迁移, 跨领域推荐

REN Hao, LIU Baisong, SUN Jinyang. Advances and Perspectives on Knowledge Transfer Based Cross-Domain Recom-mendation[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(11): 1813-1827.

任豪，刘柏嵩，孙金杨. 面向知识迁移的跨领域推荐算法研究进展[J]. 计算机科学与探索, 2020, 14(11): 1813-1827.

References

[1] Ying R, He R, Chen K, et al. Graph convolutional neural networks for web-scale recommender systems[C]//Proceed-ings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2018: 974-983.
[2] Wang D, Liang Y, Xu D, et al. A content-based recommender system for computer science publication[J]. Knowledge-Based Systems, 2018, 157: 1-9.
[3] Reddy S R S, Nalluri S, Kuniseti S, et al. Content-based movie recommendation system using genre correlation[M]//Smart Intelligent Computing and Applications. Singapore: Springer, 2019: 391-397.
[4] Lian D, Ge Y, Zhang F, et al. Scalable content-aware collabora-tive filtering for location recommendation[J]. IEEE Transac-tions on Knowledge and Data Engineering, 2018, 30(6): 1122-1135.
[5] Nilashi M, Ibrahim O, Bagherifard K. A recommender system based on collaborative filtering using ontology and dimensio-nality reduction techniques[J]. Expert Systems with Appli-catios, 2018, 92: 507-520.
[6] Fu M, Qu H, Yi Z, et al. A novel deep learning-based coll-aborative filtering model for recommendation system[J]. IEEE Transactions on Cybernetics, 2018, 49(3): 1084-1096.
[7] Xue F, He X, Wang X, et al. Deep item-based collaborative filtering for top-n recommendation[J]. ACM Transactions on Information Systems, 2019, 37(3): 1-25.
[8] Valdiviezo-Díaz P, Bobadilla J. A hybrid approach of recom-mendation via extended matrix based on collaborative filtering with demographics information[C]//Proceedings of the 2018 International Conference on Technology Trends. Cham: Sprin-ger, 2018: 384-398.
[9] Wu D. Music personalized recommendation system based on hybrid filtration[C]//Proceedings of the 2019 International Conference on Intelligent Transportation Big Data and Smart City. Piscataway: IEEE, 2019: 430-433.
[10] Dhruv A, Kamath A, Powar A, et al. Artist recommendation system using hybrid method: a novel approach[M]//Emerging Research in Computing, Information, Communication and Applications. Singapore: Springer, 2019: 527-542.
[11] Bandyopadhyay S, Thakur S S. Product prediction and recom-mendation in e-commerce using collaborative filtering and artificial neural networks: a hybrid approach[M]//Intelligent Computing Paradigm: Recent Trends. Singapore: Springer, 2020: 59-67.
[12] Veeramachaneni S D, Pujari A K, Padmanabhan V, et al. A maximum margin matrix factorization based transfer learning approach for cross-domain recommendation[J]. Applied Soft Computing, 2019, 85: 105751.
[13] Fu W, Peng Z, Wang S, et al. Deeply fusing reviews and contents for cold start users in cross-domain recommendation systems[C]//Proceedings of the 2019 AAAI Conference on Artificial Intelligence. Menlo Park: AAAI, 2019: 94-101.
[14] Fernández-Tobías I, Cantador I, Tomeo P, et al. Addressing the user cold start with cross-domain collaborative filtering: exploiting item metadata in matrix factorization[J]. User Mod-eling and User-Adapted Interaction, 2019, 29(2): 443-486.
[15] Khan M M, Ibrahim R, Ghani I. Cross domain recommender systems: a systematic literature review[J]. ACM Computing Surveys, 2017, 50(3): 1-34.
[16] Huang L, Wang C D, Chao H Y, et al. A score prediction approach for optional course recommendation via cross-user-domain collaborative filtering[J]. IEEE Access, 2019, 7: 19550-19563.
[17] Li B. Cross-domain collaborative filtering: a brief survey[C]//Proceedings of the IEEE 23rd International Conference on Tools with Artificial Intelligence. Washington: IEEE Computer Society, 2011: 1085-1086.
[18] Fernández-Tobías I, Cantador I, Kaminskas M, et al. Cross-domain recommender systems: a survey of the state of the art[C]//Spanish Conference on Information Retrieval, 2012: 24-36.
[19] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Trans-actions on Knowledge and Data Engineering, 2009, 22(10): 1345-1359.
[20] Tan C, Sun F, Kong T, et al. A survey on deep transfer lear-ning[C]//Proceedings of the 2018 International Conference on Artificial Neural Networks. Cham: Springer, 2018: 270-279.
[21] He M, Zhang J, Zhang S. ACTL: adaptive codebook transfer learning for cross-domain recommendation[J]. IEEE Access, 2019, 7: 19539-19549.
[22] Huang L, Zhao Z L, Wang C D, et al. LSCD: low-rank and sparse cross-domain recommendation[J]. Neurocomputing, 2019, 366: 86-96.
[23] Taneja A, Arora A. Cross domain recommendation using multidimensional tensor factorization[J]. Expert Systems with Applications, 2018, 92: 304-316.
[24] He M, Zhang J, Zhang J. MINDTL: multiple incomplete do-mains transfer learning for information recommendation[J]. China Communications, 2017, 14(11): 218-236.
[25] Zhuang F, Zheng J, Chen J, et al. Transfer collaborative filter-ing from multiple sources via consensus regularization[J].Neural Networks, 2018, 108: 287-295.
[26] Jiang S, Ding Z, Fu Y. Heterogeneous recommendation via deep low-rank sparse collective factorization[J]. IEEE Transac-tions on Pattern Analysis and Machine Intelligence, 2020, 42(5): 1097-1111.
[27] Wang J, Li S J, Yang S, et al. A new transfer learning model for cross-domain recommendation[J]. Chinese Journal of Com-puters, 2017, 40(10): 2367-2380.王俊, 李石君, 杨莎, 等. 一种新的用于跨领域推荐的迁移学习模型[J]. 计算机学报, 2017, 40(10): 2367-2380.
[28] Li L F, Liu Z, Wei G M, et al. Cross-domain recommenda-tion algorithm based on sharing knowledge pattern[J]. Acta Electronica Sinica, 2018, 46(8): 1947-1953.李林峰, 刘真, 魏港明, 等. 基于共享知识模型的跨领域推荐算法[J]. 电子学报, 2018, 46(8): 1947-1953.
[29] Zhang Q, Lu J, Wu D S, et al. A cross-domain recommender system with kernel-induced knowledge transfer for overlapp-ing entities[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(7): 1998-2012.
[30] Chang W, Wu Y, Liu H, et al. Cross-domain kernel induc-tion for transfer learning[C]//Proceedings of the 31st Con-ference on Artificial Intelligence. New York: ACM, 2017: 1763-1769.
[31] Zhang Q, Hao P, Lu J, et al. Cross-domain recommendation with semantic correlation in tagging systems[C]//Proceedings of the 2019 International Joint Conference on Neural Net-works. Piscataway: IEEE, 2019: 1-8.
[32] Wang X, Peng Z, Wang S, et al. Cross-domain recommenda-tion for cold-start users via neighborhood based feature map-ping[C]//Proceedings of the 2018 International Conference on Database Systems for Advanced Applications. Berlin, Heidelberg: Springer, 2018: 158-165.
[33] Wu Y W, Li B, Sun C H, et al. Research on domain adaptive recommendation methods based on transfer learning[J]. Com-puter Engineering and Applications, 2019, 55(13): 59-65.吴彦文, 李斌, 孙晨辉, 等. 基于迁移学习的领域自适应推荐方法研究[J]. 计算机工程与应用, 2019, 55(13): 59-65.
[34] Yu X, Jiang F, Du J, et al. A cross-domain collaborative filter-ing algorithm with expanding user and item features via the latent factor space of auxiliary domains[J]. Pattern Recogni-tion, 2019, 94: 96-109.
[35] Li C T, Hsu C T, Shan M K. A cross-domain recommenda-tion mechanism for cold-start users based on partial least squares regression[J]. ACM Transactions on Intelligent Sys-tems and Technology, 2018, 9(6): 1-26.
[36] Yu X, Lin J, Jiang F, et al. A cross-domain collaborative filter-ing algorithm based on feature construction and locally weighted linear regression[J]. Computational Intelligence and Neuroscience, 2018: 1-12.
[37] Ge M F, Liu Z, Wang N N, et al. Cross-domaining item recom-mendation algorithm including tag transfer[J]. Computer Science, 2019, 46(10): 1-6.葛梦凡, 刘真, 王娜娜, 等. 加入标签迁移的跨领域项目推荐算法[J]. 计算机科学, 2019, 46(10): 1-6.
[38] Ma G, Wang Y, Zheng X, et al. Leveraging transitive trust relations to improve cross-domain recommendation[J]. IEEE Access, 2018, 6: 38012-38025.
[39] Sahu A K, Dwivedi P. User profile as a bridge in cross-domain recommender systems for sparsity reduction[J]. Applied Intel-ligence, 2019, 49(7): 2461-2481.
[40] Wang Y, Feng C, Guo C, et al. Solving the sparsity problem in recommendations via cross-domain item embedding based on co-clustering[C]//Proceedings of the 12th ACM Interna-tional Conference on Web Search and Data Mining. New York: ACM, 2019: 717-725.
[41] Zhang Y, Ma X, Wan S, et al. CrossRec: cross-domain recom-mendations based on social big data and cognitive computing[J]. Mobile Networks and Applications, 2018, 23(6): 1610-1623.
[42] Shu K, Wang S, Tang J, et al. Crossfire: cross media joint friend and item recommendations[C]//Proceedings of the 11th ACM International Conference on Web Search and Data Mining. New York: ACM, 2018: 522-530.
[43] Zheng M, Bu J, Chen C, et al. Graph regularized sparse coding for image representation[J]. IEEE Transactions on Image Processing, 2010, 20(5): 1327-1336.
[44] Long M, Ding G, Wang J, et al. Transfer sparse coding for robust image representation[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recogni-tion. Washington: IEEE Computer Society, 2013: 407-414.
[45] Zhu F, Wang Y, Chen C, et al. A deep framework for cross-domain and cross-system recommendations[C]//Proceedings of the 27th International Joint Conference on Artificial Intel-ligence, Stockholm, Jul 13-19, 2018: 3711-3717.
[46] Gao C, Chen X, Feng F, et al. Cross-domain recommendation without sharing user-relevant data[C]]//Proceedings of the 28th International Conference on World Wide Web. New York: ACM, 2019: 491-502.
[47] Xu Y, Peng Z, Hu Y, et al. SARFM: a sentiment-aware review feature mapping approach for cross-domain recom-mendation[C]//Proceedings of the 2018 International Con-ference on Web Information Systems Engineering. Berlin, Heidelberg: Springer, 2018: 3-18.
[48] Kang S K, Hwang J, Lee D, et al. Semi-supervised learning for cross-domain recommendation to cold-start users[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. New York: ACM, 2019: 1563-1572.
[49] He J, Liu R, Zhuang F, et al. A general cross-domain recom-mendation framework via Bayesian neural network[C]//Pro-ceedings of the 2018 IEEE International Conference on Data Mining. Piscataway: IEEE, 2018: 1001-1006.
[50] Elkahky A M, Song Y, He X. A multi-view deep learning approach for cross domain user modeling in recommendation systems[C]//Proceedings of the 24th International Conference on World Wide Web. New York: ACM, 2015: 278-288.
[51] Hu G, Zhang Y, Yang Q. Transfer meets hybrid: a synthetic approach for cross-domain collaborative filtering with text[C]//Proceedings of the 28th International Conference on World Wide Web. New York: ACM, 2019: 2822-2829.
[52] Perera D, Zimmermann R. LSTM networks for online cross-network recommendations[C]//Proceedings of the 27th Inter-national Joint Conference on Artificial Intelligence, Stock-holm, Jul 13-19, 2018: 3825-3833.
[53] Zhao C, Li C, Fu C. Cross-domain recommendation via pre-ference propagation GraphNet[C]//Proceedings of the 28th ACM International Conference on Information and Know-ledge Management. New York: ACM, 2019: 2165-2168.
[54] Hu G, Zhang Y, Yang Q. CoNet: collaborative cross networks for cross-domain recommendation[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management. New York: ACM, 2018: 667-676.
[55] He X, Liao L, Zhang H, et al. Neural collaborative filtering[C]//Proceedings of the 26th International Conference on World Wide Web. New York: ACM, 2017: 173-182.
[56] Misra I, Shrivastava A, Gupta A, et al. Cross-stitch networks for multi-task learning[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 3994-4003.
[57] Liu J, Zhao P, Liu Y, et al. Exploiting aesthetic preference in deep cross networks for cross-domain recommendation[C]//Proceedings of the 29th International Conference on World Wide Web. New York: ACM, 2020: 1-9.
[58] Jin X, Wu L, Li X, et al. ILGNet: inception modules with connected local and global features for efficient image aes-thetic quality classification using domain adaptation[J]. IET Computer Vision, 2018, 13(2): 206-212.
[59] Wang C, Niepert M, Li H. RecSys-DAN: discriminative adversarial networks for cross-domain recommender systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(8): 2731-2740.
[60] Yuan F, Yao L, Benatallah B. DARec: deep domain adaptation for cross-domain recommendation via transferring rating pat-terns[C]//Proceedings of the 28th International Joint Con-ference on Artificial Intelligence, Macao, China, Aug 10-16, 2019: 4227-4233.
[61] Jo S Y, Jang S H, Cho H E, et al. Scenery-based fashion recommendation with cross-domain generative adversarial networks[C]//Proceedings of the 2019 IEEE International Con-ference on Big Data and Smart Computing. Piscataway: IEEE, 2019: 1-4.