融合转移关系正则化的序列推荐

doi:10.19678/j.issn.1000-3428.0069336

摘要/Abstract

摘要：

序列推荐是推荐系统中的一类重要任务, 其通过建模用户顺序行为来实现个性化、动态性的推荐。然而, 在现实环境中, 用户行为数据往往具有高度稀疏性, 同时行为序列中所包含的项目转移关系随项目特性而改变。因此, 如何充分利用用户-项目间的协同关系, 同时捕捉项目-项目间的转移规律, 成为序列推荐中至关重要的问题。针对这一问题, 提出一种融合转移关系正则化的联合矩阵分解方法。该方法通过对用户-项目交互矩阵和项目-项目间马尔可夫转移矩阵进行联合分解, 并在分解过程中设定项目表征因子共享, 共同捕捉协同关系和转移关系, 缓解用户行为数据的稀疏问题, 进而实现有效的序列推荐。在5个公开数据集上进行实验比较和分析, 结果表明, 该方法相比现有先进算法具有更好的序列推荐性能。

关键词: 序列推荐, 转移关系, 矩阵分解, 正则化, 数据稀疏

Abstract:

Sequential recommendation is a crucial task in recommendation systems for realizing personalized and dynamic recommendations by modeling users' sequential behaviors. However, in real-world scenarios, user-behavior data often exhibit a high degree of sparsity. Meanwhile, item transition relationships embedded within behavioral sequences vary depending on the item characteristics. Consequently, fully leveraging the collaborative relationships between users and items, while simultaneously capturing transition patterns among items, has emerged as a pivotal challenge in sequential recommendations. To address this issue, a joint matrix factorization method that incorporates transition relation regularization is proposed. This approach involves a joint decomposition of the user-item interaction matrix and the Markov transition matrix between items. By setting the item representation factors to be shared during the decomposition process, both collaborative and transition relationships are jointly captured, thereby alleviating the data sparsity problem associated with user behavior and enabling effective sequential recommendations. Experimental comparisons and analyses conducted on five publicly available datasets demonstrate that the proposed method outperforms existing state-of-the-art algorithms in terms of sequential recommendation.

Key words: sequential recommendation, transition relation, matrix factorization, regularization, data sparsity

冯雅莉, 温雯, 郝志峰, 蔡瑞初. 融合转移关系正则化的序列推荐[J]. 计算机工程, 2025, 51(8): 151-159.

FENG Yali, WEN Wen, HAO Zhifeng, CAI Ruichu. Sequence Recommendation with Fusion Transition Relation Regularization[J]. Computer Engineering, 2025, 51(8): 151-159.

https://www.ecice06.com/CN/Y2025/V51/I8/151

图/表 6

图1 FTRR模型的框架

Fig.1 Framework of FTRR model

图2 FTRR与变体方法的性能对比

Fig.2 Performance comparison between FTRR and variant methods

图3 FTRR的敏感性分析结果

Fig.3 Sensitivity analysis results of FTRR

参考文献 37

1	WANG S J, HU L, WANG Y, et al. Sequential recommender systems: challenges, progress and prospects[EB/OL]. [2024-01-05]. https://arxiv.org/abs/2001.04830.
2	QIU J Z, DONG Y X, MA H, et al. Network embedding as matrix factorization: unifying DeepWalk, LINE, PTE, and node2vec[C]//Proceedings of the 7th ACM International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2018: 459-467.
3	PEROZZI B, AL-RFOU R, SKIENA S. DeepWalk: online learning of social representations[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2014: 701-710.
4	GROVER A, LESKOVEC J. node2vec: scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2016: 855-864.
5	WU C Y, AHMED A, BEUTEL A, et al. Recurrent recommender networks[C]//Proceedings of the 10th ACM International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2017: 495-503.
6	ZHU Y, LI H, LIAO Y, et al. What to do next: modeling user behaviors by time-LSTM[EB/OL]. [2024-01-05]. https://dl.acm.org/doi/10.5555/3172077.3172393.
7	RENDLE S, FREUDENTHALER C, SCHMIDT-THIEME L. Factorizing personalized Markov chains for next-basket recommendation[C]//Proceedings of the 19th International Conference on World Wide Web. New York, USA: ACM Press, 2010: 811-820.
8	KANG W C, MCAULEY J. Self-attentive sequential recommendation[C]//Proceedings of the IEEE International Conference on Data Mining (ICDM). Washington D.C., USA: IEEE Press, 2018: 197-206.
9	FAN Z W, LIU Z W, WANG S, et al. Modeling sequences as distributions with uncertainty for sequential recommendation[C]//Proceedings of the 30th ACM International Conference on Information 38 Knowledge Management. New York, USA: ACM Press, 2021: 3019-3023.
10	CAI R Q, WU J B, SAN A D, et al. Category-aware collaborative sequential recommendation[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2021: 388-397.
11	DING Y J , MA Y S , WONG W K , et al. Modeling instant user intent and content-level transition for sequential fashion recommendation. IEEE Transactions on Multimedia, 2022, 24, 2687- 2700. doi: 10.1109/TMM.2021.3088281
12	TRIVEDI R, DAI H, WANG Y, et al. Know-Evolve: deep temporal reasoning for dynamic knowledge graphs[EB/OL]. [2024-01-05]. https://arxiv.org/abs/1705.05742.
13	ZHU Y C, CHEN Z Z. Mutually-regularized dual collaborative variational auto-encoder for recommendation systems[C]//Proceedings of the ACM Web Conference 2022. New York, USA: ACM Press, 2022: 2379-2387.
14	SHCHUR O, BILOŠ M, GVNNEMANN S. Intensity-free learning of temporal point processes[EB/OL]. [2024-01-05]. https://github.com/shchur/ifl-tpp.
15	CHANG X F, LIU X Q, WEN J F, et al. Continuous-time dynamic graph learning via neural interaction processes[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management. New York, USA: ACM Press, 2020: 145-154.
16	TIAN S, XIONG T, SHI L L. Streaming dynamic graph neural networks for continuous-time temporal graph modeling[C]//Proceedings of the IEEE International Conference on Data Mining (ICDM). Washington D.C., USA: IEEE Press, 2021: 1361-1366.
17	ZHANG Y, XIONG Y, LI D S, et al. CoPE: modeling continuous propagation and evolution on interaction graph[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management. New York, USA: ACM Press, 2021: 2627-2636.
18	COVINGTON P, ADAMS J, SARGIN E. Deep neural networks for YouTube recommendations[C]//Proceedings of the 10th ACM Conference on Recommender Systems. New York, USA: ACM Press, 2016: 191-198.
19	YING R, HE R N, CHEN K F, et al. Graph convolutional neural networks for web-scale recommender systems[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. New York, USA: ACM Press, 2018: 974-983.
20	张宝鑫, 杨丹, 聂铁铮, 等. 基于自监督的多视角图协同过滤推荐方法. 计算机工程, 2024, 50 (5): 100- 110. doi: 10.19678/j.issn.1000-3428.0067851
	ZHANG B X , YANG D , NIE T Z , et al. Recommendation method based on self-supervised multi-view graph collaborative filtering. Computer Engineering, 2024, 50 (5): 100- 110. doi: 10.19678/j.issn.1000-3428.0067851
21	宋玉龙, 马文明, 刘彤彤. 融合用户信任度的概率矩阵分解群组推荐算法. 计算机工程, 2022, 48 (1): 105- 111. doi: 10.19678/j.issn.1000-3428.0059526
	SONG Y L , MA W M , LIU T T . Group recommendation algorithm incorporating user trust with probability matrix factorization. Computer Engineering, 2022, 48 (1): 105- 111. doi: 10.19678/j.issn.1000-3428.0059526
22	沈学利, 马玉营, 梁振兴. 融合复杂先验与注意力机制的变分自动编码器. 计算机工程, 2022, 48 (11): 55- 61. doi: 10.19678/j.issn.1000-3428.0063401
	SHEN X L , MA Y Y , LIANG Z X . Variational auto-encoder combining complex priori and attention mechanism. Computer Engineering, 2022, 48 (11): 55- 61. doi: 10.19678/j.issn.1000-3428.0063401
23	KOREN Y , BELL R , VOLINSKY C . Matrix factorization techniques for recommender systems. Computer, 2009, 42 (8): 30- 37.
24	RENDLE S, FREUDENTHALER C, GANTNER Z, et al. BPR: Bayesian personalized ranking from implicit feedback[EB/OL]. [2024-01-05]. https://arxiv.org/abs/1205.2618.
25	HE X N, LIAO L Z, ZHANG H W, et al. Neural collaborative filtering[C]//Proceedings of the 26th International Conference on World Wide Web. New York, USA: ACM Press, 2017: 173-182.
26	TAY Y, ANH TUAN L, HUI S C. Latent relational metric learning via memory-based attention for collaborative ranking[C]//Proceedings of the 2018 World Wide Web Conference. New York, USA: ACM Press, 2018: 729-739.
27	KABBUR S, NING X, KARYPIS G. FISM: factored item similarity models for top-N recommender systems[C]//Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2013: 659-667.
28	KOREN Y. Factorization meets the neighborhood: a multifaceted collaborative filtering model[C]//Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2008: 426-434.
29	CHEN J Y, ZHANG H W, HE X N, et al. Attentive collaborative filtering: multimedia recommendation with item- and component-level attention[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2017: 335-344.
30	HE X N , HE Z K , SONG J K , et al. NAIS: neural attentive item similarity model for recommendation. IEEE Transactions on Knowledge and Data Engineering, 2018, 30 (12): 2354- 2366. doi: 10.1109/TKDE.2018.2831682
31	HU Y F, KOREN Y, VOLINSKY C. Collaborative filtering for implicit feedback datasets[C]//Proceedings of the 8th IEEE International Conference on Data Mining. Washington D.C., USA: IEEE Press, 2008: 263-272.
32	SEDHAIN S, MENON A K, SANNER S, et al. AutoRec: autoencoders meet collaborative filtering[C]//Proceedings of the 24th International Conference on World Wide Web. New York, USA: ACM Press, 2015: 111-112.
33	LIANG D W, ALTOSAAR J, CHARLIN L, et al. Factorization meets the item embedding: regularizing matrix factorization with item co-occurrence[C]//Proceedings of the 10th ACM Conference on Recommender Systems. New York, USA: ACM Press, 2016: 59-66.
34	HE X N, DENG K, WANG X, et al. LightGCN: simplifying and powering graph convolution network for recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2020: 639-648.
35	LIU J , LI D , GU H , et al. Parameter-free dynamic graph embedding for link prediction. Advances in Neural Information Processing Systems, 2022, 35, 27623- 27635.
36	LIU S J, LIU J H, GU H S, et al. AutoSeqRec: autoencoder for efficient sequential recommendation[C]//Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. New York, USA: ACM Press, 2023: 1493-1502.
37	ZHU T Y, SHI Y S, ZHANG Y, et al. Collaboration and transition: distilling item transitions into multi-query self-attention for sequential recommendation[C]//Proceedings of the 17th ACM International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2024: 1003-1011.

[1]	王帅, 史艳翠. 基于个性化数据增强的自监督序列推荐算法[J]. 计算机工程, 2025, 51(8): 190-202.
[2]	雷一凡, 陈晓红. 隐私保护的去中心联邦多视图聚类[J]. 计算机工程, 2025, 51(7): 180-189.
[3]	廖丁丁, 刘俊峰, 曾君, 邱晓欢. 一种基于块平均正交权重修正的连续学习算法[J]. 计算机工程, 2025, 51(6): 57-64.
[4]	沈忱, 何勇, 彭安浪. 鲁棒物联网多维时序数据预测方法[J]. 计算机工程, 2025, 51(4): 107-118.
[5]	林烁彬, 蔡捷仪, 方晓城, 张正, 卢光明, 陈炳志. 基于强度相关正则化学习的对抗鲁棒蒸馏方法[J]. 计算机工程, 2025, 51(1): 42-50.
[6]	郭敏, 张熙涵, 李阳. 融合注意力的教师互一致性半监督医学图像分割[J]. 计算机工程, 2024, 50(9): 313-323.
[7]	李忠伟, 周洁, 刘昕, 吴金燠, 李可一. 融合时间和知识信息的生成对抗网络序列推荐算法[J]. 计算机工程, 2024, 50(11): 70-79.
[8]	陈君航, 杨祖元, 刘名扬, 李陵江. 基于正交约束的广义可分离非负矩阵分解算法[J]. 计算机工程, 2023, 49(8): 46-53.
[9]	冉瑞生, 翁稳稳, 王宁, 彭顺顺. 基于人脸关键特征提取的表情识别[J]. 计算机工程, 2023, 49(2): 254-262.
[10]	李盼, 解庆, 李琳, 刘永坚. 知识增强的图神经网络序列推荐模型[J]. 计算机工程, 2023, 49(2): 70-80.
[11]	雷洁, 饶文碧, 杨焱超, 熊盛武. 基于分类不确定性的伪标签目标检测算法[J]. 计算机工程, 2023, 49(1): 49-56.
[12]	刘杭, 殷歆, 陈杰, 罗恒. 基于混合网络模型的多维时间序列预测[J]. 计算机工程, 2023, 49(1): 121-129.
[13]	范林歌, 武欣嵘, 童玮, 曾维军. 基于概率矩阵分解的不完整数据集特征选择方法[J]. 计算机工程, 2022, 48(6): 57-64.
[14]	冉懿, 王润年, 潘红伟, 俞海猛, 袁培森. 面向停电分类预测的因子分解机模型[J]. 计算机工程, 2022, 48(5): 98-103,111.
[15]	王丽娟, 张霖, 尹明, 郝志峰, 蔡瑞初, 温雯. 基于正交基的多视图迁移谱聚类[J]. 计算机工程, 2022, 48(10): 37-44,54.

选择文件类型/文献管理软件名称

选择包含的内容