Survey on Solving Cold Start Problem in Recommendation Systems

doi:10.3778/j.issn.1673-9418.2308044

Abstract

Abstract: Recommender systems provide important functions in areas such as dealing with data overload, providing personalized consulting services, and assisting clients in investment decisions. However, the cold start problem in recommender systems has always been in urgent need of solution and optimization. Based on this, this paper classifies the traditional methods and cutting-edge methods to solve the cold start problem, and expounds the research progress and excellent methods in recent years. Firstly, three traditional solutions to the cold start problem are summarized: recommendation based on content filtering, recommendation based on collaborative filtering, and hybrid recommendation. Secondly, the current cutting-edge recommendation algorithms to solve the cold start problem are summarized, and they are classified into the data-driven strategy and the method-driven strategy. The method-driven strategy is divided into algorithms based on meta-learning, algorithms based on context information and session str-ategy, algorithms based on random walk, algorithms based on heterogeneous graph information and attribute graph, and algorithms based on adversarial mechanism. According to the type of cold start problem, the algorithms are divided into two categories: new users and new items. Then, according to the particularity of the recommendation field, the cold start problem of the recommendation in the multimedia information field and the online e-commerce platform field is expounded. Finally, the possible research directions to solve the cold start problem in the future are summarized.

Key words: cold start, recommender systems, meta-learning, context information, random walk

摘要： 推荐系统在处理数据超载、提供个性化咨询服务、帮助客户投资决策等领域提供了重要功能。但推荐系统中存在的冷启动问题一直亟需解决和优化。基于此，对解决冷启动问题的传统方法和前沿方法进行分类，将近几年的研究进展和优秀的方法进行阐述。首先，归纳了冷启动问题的传统三大解决方案：基于内容过滤的推荐、基于协同过滤的推荐和混合推荐。其次，归纳了目前较为前沿的解决冷启动的推荐算法，并依据其解决冷启动问题的策略点将其分类为数据驱动的策略和方法驱动的策略，再将方法驱动的策略分为基于元学习的算法、基于上下文信息和会话策略的算法、基于随机游走的算法、基于异质图信息和属性图的算法和基于对抗性机制的算法，其中根据处理冷启动问题的种类将算法分为解决新用户和新项目两类。再根据推荐领域的特殊性，将多媒体信息领域推荐和在线电商平台领域推荐的冷启动问题进行阐述。最后，总结并提出了未来解决冷启动问题可能的研究方向。

关键词: 冷启动, 推荐系统, 元学习, 上下文信息, 随机游走

MAO Qian, XIE Weicheng, QIAO Yitian, HUANG Xiaolong, DONG Gang. Survey on Solving Cold Start Problem in Recommendation Systems[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(5): 1197-1210.

毛骞, 谢维成, 乔逸天, 黄小龙, 董刚. 推荐系统冷启动问题解决方法研究综述[J]. 计算机科学与探索, 2024, 18(5): 1197-1210.

References

[1] CAMACHO L A G, ALVES-SOUZA S N. Social network data to alleviate cold-start in recommender system: a systematic review[J]. Information Processing & Management, 2018, 54(4): 529-544.
[2] DAFALLA A, WEI L, LIAO Z, et al. Effects of clamping pressure on cold start behavior of polymer electrolyte fuel cells[J]. Fuel Cells, 2019, 19(3): 221-230.
[3] 于蒙, 何文涛, 周绪川, 等. 推荐系统综述[J]. 计算机应用, 2022, 42(6): 1898-1913.
YU M, HE W T, ZHOU X C, et al. Review of recommendation system[J]. Journal of Computer Applications, 2022, 42(6): 1898-1913.
[4] WEI J, HE J, CHEN K, et al. Collaborative filtering and deep learning based recommendation system for cold start items[J]. Expert Systems with Applications, 2017, 69: 29-39.
[5] JEEVAMOL J, RENUMOL V. An ontology-based hybrid e-learning content recommender system for alleviating the cold-start problem[J]. Education and Information Technologies, 2021, 26: 4993-5022.
[6] ANWAAR F, ILTAF N, AFZAL H, et al. HRS-CE: a hybrid framework to integrate content embeddings in recommender systems for cold start items[J]. Journal of Computational Science, 2018, 29: 9-18.
[7] ZHANG Y, SHI Z, ZUO W, et al. Joint personalized Markov chains with social network embedding for cold-start recommendation[J]. Neurocomputing, 2020, 386: 208-220.
[8] ZHANG Z, ZHANG Y, REN Y. Employing neighborhood reduction for alleviating sparsity and cold start problems in user-based collaborative filtering[J]. Information Retrieval Journal, 2020, 23: 449-472.
[9] 范永全, 杜亚军, 成丽静, 等. 一种融合用户偏好与信任度的增强协同过滤推荐方法[J]. 西华大学学报(自然科学版), 2015, 34(4): 8-12.
FAN Y Q, DU Y J, CHENG L J, et al. An improved collaborative filtering recommendation method with user trust-preference fusion[J]. Journal of Xihua University (Natural Science) , 2015, 34(4): 8-12.
[10] 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.
[11] HU J, LIU L, ZHANG C, et al. Hybrid recommendation algorithm based on latent factor model and PersonalRank[J]. Journal of Internet Technology, 2018, 19(3): 919-926.
[12] FENG J, XIA Z, FENG X, et al. RBPR: a hybrid model for the new user cold start problem in recommender systems [J]. Knowledge-Based Systems, 2021, 214: 106732.
[13] 于旭, 何亚东, 杜军威, 等. 一种结合显式特征和隐式特征的开发者混合推荐算法[J]. 软件学报, 2022, 33(5): 1635-1651.
YU X, HE Y D, DU J W, et al. Developer hybrid recommendation algorithm based on combination of explicit features and implicit features[J]. Journal of Software, 2022, 33(5): 1635-1651.
[14] HERCE-ZELAYA J, PORCEL C, TEJEDA-LORENTE á, et al. Introducing CSP dataset: a dataset optimized for the study of the cold start problem in recommender systems [J]. Information, 2022, 14(1): 19.
[15] MU R, ZENG X. Auxiliary stacked denoising autoencoder based collaborative filtering recommendation[J]. KSII Tran-sactions on Internet and Information Systems, 2020, 14(6): 2310-2332.
[16] RAIGOZA J, KARANDE V. A study and implementation of a movie recommendation system in a cloud-based environment[J]. International Journal of Grid and High Performance Computing, 2017, 9(1): 25-36.
[17] LIKA B, KOLOMVATSOS K, HADJIEFTHYMIADES S. Facing the cold start problem in recommender systems[J]. Expert Systems with Applications, 2014, 41(4): 2065-2073.
[18] XU C. A novel recommendation method based on social network using matrix factorization technique[J]. Information Processing & Management, 2018, 54(3): 463-474.
[19] LI B. Cross-domain collaborative filtering: a brief survey[C]//Proceedings of the 2011 IEEE 23rd International Conference on Tools with Artificial Intelligence. Washington:IEEE Computer Society, 2011: 1085-1086.
[20] 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.
REN H, LIU B S, SUN J Y. Advances and perspectives on knowledge transfer based cross-domain recommendation[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(11): 1813-1827.
[21] WANG X, PENG Z, WANG S, et al. CDLFM: cross-domain recommendation for cold-start users via latent feature map-ping[J]. Knowledge and Information Systems, 2020, 62: 1723-1750.
[22] CAO J, SHENG J, CONG X, et al. Cross-domain recommendation to cold-start users via variational information bottleneck[C]//Proceedings of the 2022 IEEE 38th International Conference on Data Engineering. Piscataway: IEEE, 2022: 2209-2223.
[23] KRISHNAN A, DAS M, BENDRE M, et al. Transfer learning via contextual invariants for one-to-many cross-domain recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 1081-1090.
[24] STRUB F, MARY J, GAUDEL R. Hybrid collaborative filtering with autoencoders[J]. arXiv:1603.00806, 2016.
[25] LIU H, WANG L, LI P, et al. Relation-propagation meta-learning on an explicit preference graph for cold-start recommendation[J]. Knowledge-Based Systems, 2023, 272: 110579.
[26] DU Y, ZHU X, CHEN L, et al. MetaKG: meta-learning on knowledge graph for cold-start recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2023, 35(10): 9850-9863.
[27] WANG H, ZHAO Y. ML2E: meta-learning embedding ensemble for cold-start recommendation[J]. IEEE Access, 2020, 8: 165757-165768.
[28] KIM M, YANG Y, RYU J H, et al. Meta-learning with adaptive weighted loss for imbalanced cold-start recommendation[J]. arXiv:2302.14640, 2023.
[29] SHU H, CHUNG F L, LIN D. MetaGC-MC: a graph-based meta-learning approach to cold-start recommendation with/without auxiliary information[J]. Information Sciences, 2023, 623: 791-811.
[30] SUN Z, YANG J, ZHANG J, et al. Recurrent knowledge graph embedding for effective recommendation[C]//Proceedings of the 12th ACM Conference on Recommender Systems. New York: ACM, 2018: 297-305.
[31] WANG X, HE X, CAO Y, et al. KGAT: knowledge graph attention network for recommendation[C]//Proceedings of the 25th ACM SIGKDD International Conference on Know-ledge Discovery and Data Mining. New York: ACM, 2019: 950-958.
[32] SONG W, XIAO Z, WANG Y, et al. Session-based social recommendation via dynamic graph attention networks[C]//Proceedings of the 12th ACM International Conference on Web Search and Data Mining. New York: ACM, 2019: 555-563.
[33] 王美玲, 刘晓楠, 尹美娟, 等. 基于评论和物品描述的深度学习推荐算法[J]. 计算机科学, 2022, 49(3): 99-104.
WANG M L, LIU X N, YIN M J, et al. Deep learning recommendation algorithm based on reviews and item descriptions[J]. Computer Science, 2022, 49(3): 99-104.
[34] CHEN T, WONG R C W. Handling information loss of graph neural networks for session-based recommendation [C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2020: 1172-1180.
[35] XU C, ZHAO P, LIU Y, et al. Graph contextualized self- attention network for session-based recommendation[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence, Macao, China, Aug 10-16, 2019: 3940- 3946.
[36] WANG Z, WEI W, CONG G, et al. Global context enhanced graph neural networks for session-based recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 169-178.
[37] UR REHMAN I, ALI W, JAN Z, et al. CAML: contextual augmented meta-learning for cold-start recommendation[J]. Neurocomputing, 2023, 533: 178-190.
[38] 孙爱晶, 王国庆. 邻居关系感知的图卷积网络推荐模型 [J]. 计算机工程与应用, 2023, 59(9): 112-122.
SUN A J, WANG G Q. Neighbor relation-aware graph convolutional network for recommendation[J]. Computer Engineering and Applications, 2023, 59(9): 112-122.
[39] 曹阳, 高旻, 余俊良, 等. 基于双图混合随机游走的社会化推荐模型[J]. 电子学报, 2023, 51(2): 286-296.
CAO Y, GAO M, YU J L, et al. Bi-graph mix-random walk based social recommendation model[J]. Acta Electronica Sinica, 2023, 51(2): 286-296.
[40] WANG J, MEI H, LI K, et al. Collaborative filtering model of graph neural network based on random walk[J]. Applied Sciences, 2023, 13(3): 1786.
[41] 耿秀丽, 牛璐. 基于不同搜索路径下成对随机游走的推荐算法[J/OL]. 计算机集成制造系统 [2023-07-20]. http://kns.cnki.net/kcms/detail/11.5946.TP.20211112.1135.006.html.
GENG X L, NIU L. Recommendation algorithm based on pairwise random walk under different search paths[J/OL]. Computer Integrated Manufacturing Systems [2023-07-20]. http://kns.cnki.net/kcms/detail/11.5946.TP.20211112.1135.006.html.
[42] QIAN T, LIANG Y, LI Q, et al. Attribute graph neural networks for strict cold start recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(8): 3597-3610.
[43] 吴正洋, 张广涛, 黄立, 等. 基于异质图嵌入和会话交互的课程推荐模型[J/OL]. 计算机工程 [2023-07-20]. https://doi.org/10.19678/j.issn.1000-3428.0067554.
WU Z Y, ZHANG G T, HUANG L, et al. Course recommendation based on heterogeneous graph embedding and session interaction[J/OL]. Computer Engineering [2023-07-20]. https://doi.org/10.19678/j.issn.1000-3428.0067554.
[44] 葛尧, 陈松灿. 面向推荐系统的图卷积网络[J]. 软件学报, 2020, 31(4): 1101-1112.
GE Y, CHEN S C. Graph convolutional network for recommender systems[J]. Journal of Software, 2020, 31(4): 1101-1112.
[45] 高巍, 朱风兰, 李大舟, 等. 图神经网络在冷启动推荐中的实现[J]. 计算机工程与设计, 2022, 43(9): 2557-2566.
GAO W, ZHU F L, LI D Z, et al. Realization of graph neural network in cold start recommendation[J]. Computer Engineering and Design, 2022, 43(9): 2557-2566.
[46] 李祥霞, 陈楷锐, 李彬. 融合多维梯度反馈的生成对抗网络推荐系统[J/OL]. 计算机科学与探索 [2023-07-20]. http://kns.cnki.net/kcms/detail/11.5602.TP.20230705.0952.002.html.
LI X X, CHEN K R, LI B. Generative adversarial network recommendation system with multi-dimensional gradient feedback mechanism[J]. Journal of Frontiers of Computer Science and Technology [2023-07-20]. http://kns.cnki.net/ kcms/detail/11.5602.TP.20230705.0952.002.html.
[47] XIE Z, LIU C, ZHANG Y, et al. Adversarial and contrastive variational autoencoder for sequential recommendation[C]//Proceedings of the Web Conference 2021. New York: ACM, 2021: 449-459.
[48] CHEN C C, LAI P L, CHEN C Y. ColdGAN: an effective cold-start recommendation system for new users based on generative adversarial networks[J]. Applied Intelligence, 2023, 53(7): 8302-8317.
[49] WU H, LONG J, LI N, et al. Adversarial auto-encoder domain adaptation for cold-start recommendation with positive and negative hypergraphs[J]. ACM Transactions on Information Systems, 2022, 41(2): 1-25.
[50] WANG H, ZHANG F, ZHAO M, et al. Multi-task feature learning for knowledge graph enhanced recommendation [C]//Proceedings of the 2019 World Wide Web Conference. New York: ACM, 2019: 2000-2010.
[51] 高仰, 刘渊. 融合社交关系和知识图谱的推荐算法[J]. 计算机科学与探索, 2023, 17(1): 238-250.
GAO Y, LIU Y. Recommendation algorithm combining social relationship and knowledge graph[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(1): 238-250.
[52] ZHAO W, LI Y, FAN T, et al. A novel embedding learning framework for relation completion and recommendation based on graph neural network and multi-task learning[J/OL]. Soft Computing [2023-07-20]. https://doi.org/10.1007/s00500- 021-06617-0.
[53] 梅雨竹, 胡竹林, 朱欣娟. 融合双层注意力机制的群组偏好融合策略研究[J]. 计算机工程与应用, 2023, 59(9): 272-279.
MEI Y Z, HU Z L, ZHU X J. Research on group preference fusion strategy based on two-layer attention mechanism[J]. Computer Engineering and Applications, 2023, 59(9): 272-279.
[54] 王利娥, 李东城, 李先贤. 基于跨域关联与隐私保护的深度推荐模型[J]. 软件学报, 2023, 34(7): 3365-3384.
WANG L E, LI D C, LI X X. Deep recommendation model with cross-domain association and privacy protection[J]. Journal of Software, 2023, 34(7): 3365-3384.
[55] ZHOU M, ZHANG C, HAN X, et al. Knowledge graph completion for hyper-relational data[C]//Proceedings of the 2nd International Conference on Big Data Computing and Communications, Shenyang, Jul 29-31, 2016. Cham: Springer, 2016: 236-246.
[56] SHEN T, JIA J, LI Y, et al. PEIA: personality and emotion integrated attentive model for music recommendation on social media platforms[C]//Proceedings of the 2020 AAAI Conference on Artificial Intelligence. Menlo Park: AAAI, 2020: 206-213.
[57] MAGRON P, FéVOTTE C. Neural content-aware collaborative filtering for cold-start music recommendation[J]. Data Mining and Knowledge Discovery, 2022, 36(5): 1971-2005.
[58] LI C, LI Y, WANG C, et al. The multimedia recommendation algorithm based on probability graphical model[J]. Multimedia Tools and Applications, 2022, 81(14): 19035-19050.
[59] ZHANG Z, SUN R, CHOO K K R, et al. A novel social situation analytics-based recommendation algorithm for multimedia social networks[J]. IEEE Access, 2019, 7: 117749-117760.
[60] LA GATTA V, MOSCATO V, PENNONE M, et al. Music recommendation via hypergraph embedding[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(10): 7887-7899.
[61] HUNG T Y, HUANG S H. Addressing the cold-start problem of recommendation systems for financial products by using few-shot deep learning[J]. Applied Intelligence, 2022, 52(13): 15529-15546.
[62] VIKTORATOS I, TSADIRAS A. A machine learning approach for solving the frozen user cold-start problem in personalized mobile advertising systems[J]. Algorithms, 2022, 15(3): 72.
[63] LI Z, AMAGATA D, ZHANG Y, et al. HML4Rec: hierarchical meta-learning for cold-start recommendation in flash sale e-commerce[J]. Knowledge-Based Systems, 2022, 255: 109674.
[64] 程秀峰, 张孜铭. 基于情境感知的电商平台推荐系统框架研究[J]. 情报理论与实践, 2021, 44(2): 168-177.
CHEN X F, ZHANG Z M. The framework of E-commerce platform recommendation system based on context-awareness[J]. Information Studies: Theory & Application, 2021, 44(2): 168-177.
[65] HENK V, VAHDATI S, NAYYERI M, et al. Metaresearch recommendations using knowledge graph embeddings[C]//Proceedings of the 2019 Workshop on Recommender Systems and Natural Language Processing, Honolulu, Jan 27-28, 2019. Menlo Park: AAAI, 2019: 1-6.
[66] CHEN W, ZHANG Y, XIAN Y, et al. Hotspot information network and domain knowledge graph aggregation in heterogeneous network for literature recommendation[J]. Applied Sciences, 2023, 13(2): 1093.
[67] MANJU G, ABHINAYA P, HEMALATHA M, et al. Cold start problem alleviation in a research paper recommendation system using the random walk approach on a heterogeneous user-paper graph[J]. International Journal of Intelligent Information Technologies, 2020, 16(2): 24-48.
[68] 刘振鹏, 尹文召, 王文胜, 等. HRS-DC: 基于深度学习的混合推荐模型[J]. 计算机工程与应用, 2020, 56(14): 169-175.
LIU Z P, YIN W Z, WANG W S, et al. HRS-DC: hybrid recommendation model based on deep learning[J]. Computer Engineering and Applications, 2020, 56(14): 169-175.
[69] 付峻宇, 朱小栋, 陈晨. 基于图卷积的双通道协同过滤推荐算法[J]. 计算机应用研究, 2023, 40(1): 129-135.
FU J Y, ZHU X D, CHEN C. Two-channel collaborative filtering recommendation algorithm based on graph convolution[J]. Application Research of Computers, 2023, 40(1): 129-135.