Multi-level User Interest and Multi-intent Fusion for Next Basket Recommendation

doi:10.3778/j.issn.1673-9418.2404003

Abstract

Abstract: Next basket recommendation aims to recommend the next basket of items that users may be interested in based on the basket sequence of user historical interactions. The existing next-basket recommendation algorithms fail to adequately disentangle multi-intents within baskets and consider user interests or intents from only a single level, thereby resulting in suboptimal recommendation performance. To address the limitations, this paper proposes a multi-level user interest and multi-intent fusion model (MLIMI) for next-basket recommendation. This model separately considers user interests and multi-intents from multiple levels. Firstly, a global-level user-item interaction graph is constructed. Considering that user behavior changes over time, a long and short-term time decay weight is designed to balance the importance of the interaction items, and then the user’s dynamic interests are learnt through graph convolution networks. Secondly, a local-level basket-item graph is constructed to learn the disentangled multi-intents within baskets via a graph disentangled network, and subsequently the multi-intents are encoded via a multi-head self-attention layer to obtain the final intent representations. A cross-level contrastive learning paradigm is also designed to combine item representations from different levels in order to enhance the semantic information between items at different levels. Finally, user interests and intents from different levels are fused in the predict layer for the next basket of predictions. Comparative experiments with mainstream models such as MITGNN, TAIW and MINN on two public benchmark datasets, TaFeng and Dunnhumby, show that MLIMI outperforms many current baseline models.

Key words: next basket recommendation, graph disentangled network, multi-intent learning, contrastive learning, multi-head attention mechanism

摘要： 下一篮推荐旨在根据用户历史交互的篮子序列，为用户推荐下一篮可能感兴趣的商品。针对现有下一篮推荐算法未能较好解离篮子内的多意图以及仅从单一层面考虑用户的兴趣或意图，导致推荐效果受限等问题，提出了一种多层级用户兴趣与多意图融合的下一篮推荐模型（MLIMI），从多个层级分别考虑用户兴趣与多意图，构建全局级的用户-项目交互图。考虑到用户行为会随时间发生变化，设计一种长短期时间衰减权重平衡交互项的重要性，通过图卷积网络学习用户的动态兴趣；构建局部级篮子-项目图，通过图解离网络学习解离化的篮子内多意图，随后通过一个多头自注意力层对多意图进行编码，得到最终的意图表示。设计一个跨层级的对比学习范式，结合来自不同层级的项目表示，以增强不同层级项目之间的语义信息。在预测层中融合来自不同层级的用户兴趣和意图，进行下一篮预测。在两个公共基准数据集TaFeng和Dunnhumby上与MITGNN、TAIW、MINN等主流模型进行了对比实验，结果表明MLIMI的性能优于当前许多基线模型。

WEI Chuyuan, YUAN Baojie, WANG Changdong. Multi-level User Interest and Multi-intent Fusion for Next Basket Recommendation[J]. Journal of Frontiers of Computer Science and Technology, 2025, 19(3): 749-763.

魏楚元, 袁保杰, 王昌栋. 多层级用户兴趣与多意图融合的下一篮推荐算法[J]. 计算机科学与探索, 2025, 19(3): 749-763.

References

[1] 吴正洋, 汤庸, 刘海. 个性化学习推荐研究综述[J]. 计算机科学与探索, 2022, 16(1): 21-40.
WU Z Y, TANG Y, LIU H. Survey of personalized learning recommendation[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(1): 21-40.
[2] RICCI F, ROKACH L, SHAPIRA B. Recommender systems: techniques, applications, and challenges[M]//Recommender systems handbook. New York: Springer US, 2021: 1-35.
[3] 蒋滨泽, 邓欣, 杜雨露, 等. 基于物品关联协同过滤的下一购物篮推荐算法[J]. 计算机科学, 2023, 50(S2): 474-479.
JIANG B Z, DENG X, DU Y L, et al. Next-basket recommendation algorithm based on correlation between items collaborative filtering[J]. Computer Science, 2023, 50(S2):474-479.
[4] 周洋涛, 褚华, 杨文勇, 等. 基于时间感知与协同挖掘的下一购物篮推荐方法[J]. 武汉大学学报(理学版), 2021, 67(6): 525-531.
ZHOU Y T, CHU H, YANG W Y, et al. Next-basket recommendation based on time perception and collaborative mining[J]. Journal of Wuhan University (Natural Science Edition), 2021, 67(6): 525-531.
[5] 余文婷, 吴云. 时间感知的双塔型自注意力序列推荐模型[J]. 计算机科学与探索, 2024, 18(1): 175-188.
YU W T, WU Y. Time-aware sequential recommendation model based on dual-tower self-attention[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(1): 175-188.
[6] KANG W C, MCAULEY J. Self-attentive sequential recommendation[C]//Proceedings of the 2018 IEEE International Conference on Data Mining. Piscataway: IEEE, 2018: 197-206.
[7] 吴文政, 卢先领. 融合物品转换关系和时序信息的会话推荐算法[J]. 计算机科学与探索, 2024, 18(3): 768-779.
WU W Z, LU X L. Session recommendation algorithm combining item transition relations and time-order information[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(3): 768-779.
[8] WU S, TANG Y Y, ZHU Y Q, et al. Session-based recommendation with graph neural networks[C]//Proceedings of the 33rd AAAI Conference on Artificial Intelligence and the 31st Innovative Applications of Artificial Intelligence Conference and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence. Palo Alto: AAAI, 2019: 346-353.
[9] WANG S J, HU L, WANG Y, et al. Intention nets: psychology-inspired user choice behavior modeling for next-basket prediction[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(4): 6259-6266.
[10] RENDLE S, FREUDENTHALER C, SCHMIDT-THIEME L, et al. Factorizing personalized Markov chains for next-basket recommendation[C]//Proceedings of the 19th International Conference on World Wide Web. New York: ACM, 2010: 811-820.
[11] BALTRUNAS L, LUDWIG B, RICCI F, et al. Matrix factorization techniques for context aware recommendation[C]//Proceedings of the 5th ACM Conference on Recommender Systems. New York: ACM, 2011: 301-304.
[12] LENG Y F, YU L, XIONG J, et al. Recurrent convolution basket map for diversity next-basket recommendation[M]//Database systems for advanced applications. Cham: Springer, 2020: 638-653.
[13] CHEN Y J, LI J, LIU C H, et al. Modeling dynamic attributes for next basket recommendation[EB/OL]. [2024-02-13]. https://arxiv.org/abs/2109.11654.
[14] LE D T, LAUW H W, FANG Y, et al. Correlation-sensitive next-basket recommendation[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence, 2019: 2808-2814.
[15] YU L, SUN L L, DU B W, et al. Predicting temporal sets with deep neural networks[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2020: 1083-1091.
[16] WEI J, HE J H, 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.
[17] YU F, LIU Q, WU S, et al. A dynamic recurrent model for next basket recommendation[C]//Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2016: 729-732.
[18] HU H J, HE X N. Sets2Sets: learning from sequential sets with neural networks[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2019: 1491-1499.
[19] ARIANNEZHAD M, JULLIEN S, LI M, et al. ReCANet: a repeat consumption-aware neural network for next basket recommendation in grocery shopping[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2022: 1240-1250.
[20] YU Y, SI X S, HU C H, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235-1270.
[21] DEY R, SALEM F M. Gate-variants of gated recurrent unit (GRU) neural networks[C]//Proceedings of the 2017 IEEE 60th International Midwest Symposium on Circuits and Systems. Piscataway: IEEE, 2017: 1597-1600.
[22] BAI T, NIE J Y, ZHAO W X, et al. An attribute-aware neural attentive model for next basket recommendation[C]//Proceedings of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2018: 1201-1204.
[23] CHE B B, ZHAO P P, FANG J H, et al. Inter-basket and intra- basket adaptive attention network for next basket recommendation[J]. IEEE Access, 2019, 7: 80644-80650.
[24] SONG T S, GUO F, JIANG H R, et al. HGAT-BR: hyperedge-based graph attention network for basket recommendation[J]. Applied Intelligence, 2023, 53(2): 1435-1451.
[25] YUAN W H, WANG H, YU X M, et al. Attention-based context-aware sequential recommendation model[J]. Information Sciences, 2020, 510: 122-134.
[26] ZHOU J, CUI G Q, HU S D, et al. Graph neural networks: a review of methods and applications[J]. AI Open, 2020, 1: 57-81.
[27] ZHANG Y, GUO B, WANG Q R, et al. MGCN4REC: multi-graph convolutional network for next basket recommendation with instant interest[M]//Green, pervasive, and cloud computing. Cham: Springer, 2020: 171-185.
[28] SU T T, WANG C D, XI W D, et al. Hierarchical alignment with polar contrastive learning for next-basket recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(1): 199-210.
[29] WANG H W, ZHAO M, XIE X, et al. Knowledge graph convolutional networks for recommender systems[C]//Proceedings of the 2019 World Wide Web Conference. New York: ACM, 2019: 3307-3313.
[30] FAN Z W, LIU Z W, ZHANG J W, et al. Continuous-time sequential recommendation with temporal graph collaborative transformer[C]//Proceedings of the 30th ACM International Conference on Information and Knowledge Management. New York: ACM, 2021: 433-442.
[31] HE X N, DENG K, WANG X, et al. LightGCN[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 639-648.
[32] YU Y L, YANG E N, GUO G B, et al. Basket representation learning by hypergraph convolution on repeated items for next-basket recommendation[C]//Proceedings of the 32nd International Joint Conference on Artificial Intelligence, 2023: 2415-2422.
[33] LI R, ZHANG L, LIU G N, et al. Next basket recommendation with intent-aware hypergraph adversarial network[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2023: 1303-1312.
[34] GAO R, TAO Y H, YU Y H, et al. Self-supervised dual hypergraph learning with intent disentanglement for session-based recommendation[J]. Knowledge-Based Systems, 2023, 270: 110528.
[35] HU H J, HE X N, GAO J Y, et al. Modeling personalized item frequency information for next-basket recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 1071-1080.
[36] WAN S, LAN Y, WANG P, et al. Next basket recommendation with neural networks[C]//Poster Proceedings of the 9th ACM Conference on Recommender Systems. New York: ACM, 2015: 268-273.
[37] QIN Y Q, WANG P F, LI C L, et al. The world is binary: contrastive learning for denoising next basket recommendation[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2021: 859-868.
[38] ROMANOV A, LASHININ O, ANANYEVA M, et al. Time-aware item weighting for the next basket recommendations[C]//Proceedings of the 17th ACM Conference on Recommender Systems. New York: ACM, 2023: 985-992.
[39] CHOU Y H, CHENG P J, CHOU Y H, et al. Incorporating co-purchase correlation for next-basket recommendation[C]//Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. New York: ACM, 2023: 3823-3827.
[40] LIU T, YIN X R, NI W J. Next basket recommendation model based on attribute-aware multi-level attention[J]. IEEE Access, 2020, 8: 153872-153880.
[41] WANG X, HE X N, CAO Y X, et al. KGAT[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2019: 950-958.
[42] 潘春雨, 赵朋朋. 基于图注意力网络的下一个购物篮推荐[J]. 计算机应用与软件, 2023, 40(8): 17-23.
PAN C Y, ZHAO P P. Graph attention network for next-basket recommendation[J]. Computer Applications and Software, 2023, 40(8): 17-23.
[43] LIU T, LIU B J. Next basket recommendation based on graph attention network and transformer[J]. Journal of Physics: Conference Series, 2022(1): 012023.
[44] WASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30, 2017: 5998-6008.
[45] HE X R, WEI T X, HE J R, et al. Robust basket recommendation via noise-tolerated graph contrastive learning[C]//Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. New York: ACM, 2023: 709-719.
[46] MA J, CUI P, KUANG K, et al. Disentangled graph convolutional networks[C]//Proceedings of the 36th International Conference on Machine Learning. New York: ACM, 2019: 4212-4221.
[47] CHEN H, DENG Y J, LI Y F, et al. RGBD salient object detection via disentangled cross-modal fusion[J]. IEEE Transactions on Image Processing, 2020, 29: 8407-8416.
[48] JOHN V, MOU L L, BAHULEYAN H, et al. Disentangled representation learning for non-parallel text style transfer[EB/OL]. [2024-02-13]. https://arxiv.org/abs/1808.04339.
[49] MA J, ZHOU C, CUI P, et al. Learning disentangled representations for recommendation[C]//Advances in Neural Information Processing Systems 32, 2019: 5711-5722.
[50] HIGGINS I, MATTHEY L, PAL A, et al. Beta-VAE: learning basic visual concepts with a constrained variational framework[C]//Proceedings of the 5th International Conference on Learning Representations, 2017: 427-441.
[51] WANG X, JIN H Y, ZHANG A, et al. Disentangled graph collaborative filtering[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 1001-1010.
[52] WU J H, FAN W Q, CHEN J F, et al. Disentangled contrastive learning for social recommendation[C]//Proceedings of the 31st ACM International Conference on Information and Knowledge Management. New York: ACM, 2022: 4570-4574.
[53] WANG Y F, TANG S Y, LEI Y T, et al. DisenHAN: disentangled heterogeneous graph attention network for recommendation[C]//Proceedings of the 29th ACM International Conference on Information and Knowledge Management. New York: ACM, 2020: 1605-1614.
[54] 徐雪东, 刘晓东. 基于时间加权ALS模型协同过滤推荐算法[J]. 电子设计工程, 2022, 30(14): 39-43.
XU X D, LIU X D. Collaborative filtering recommendation algorithm based on time weighted ALS model[J]. Electronic Design Engineering, 2022, 30(14): 39-43.
[55] 殷佳莉, 江智威, 杨毅, 等. 融合时间衰减函数的改进协同过滤算法[J]. 计算机技术与发展, 2022, 32(4): 170-174.
YIN J L, JIANG Z W, YANG Y, et al. An improved collaborative filtering algorithm incorporating time decay function[J]. Computer Technology and Development, 2022, 32(4): 170-174.
[56] WANG P F, GUO J F, LAN Y Y, et al. Learning hierarchical representation model for NextBasket recommendation[C]//Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2015: 403-412.
[57] SU T T, HE Z Y, CHEN M S, et al. Basket booster for prototype-based contrastive learning in next basket recommendation[M]//Machine learning and knowledge discovery in databases. Cham: Springer, 2023: 574-589.
[58] HIDASI B, KARATZOGLOU A, BALTRUNAS L, et al. Session-based recommendations with recurrent neural networks[EB/OL]. [2024-02-13]. https://arxiv.org/abs/1511.06939.
[59] LIU Z W, LI X H, FAN Z W, et al. Basket recommendation with multi-intent translation graph neural network[C]//Proceedings of the 2020 IEEE International Conference on Big Data. Piscataway: IEEE, 2020: 728-737.
[60] DENG Z Y, LI J J, GUO Z Q, et al. Multi-aspect interest neighbor-augmented network for next-basket recommendation[C]//Proceedings of the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway: IEEE, 2023: 1-5.