基于全局图和多粒度意图单元的会话推荐

doi:10.19678/j.issn.1000-3428.0065499

摘要/Abstract

摘要：

现有基于图神经网络的会话推荐模型通过捕获项目复杂转换模式挖掘项目之间的潜在信息，但极少考虑跨会话信息及当前会话中的高层次信息，因此无法捕捉会话中复杂的依赖关系。针对该问题，建立基于全局图和多粒度意图单元的会话推荐模型。构造跨会话图，利用图注意力网络得到跨会话表示。在连续意图单元上，构建多粒度意图单元异构会话图，得到全局和局部表示。将跨会话、全局和局部表示进行融合，捕捉会话中项目之间的复杂依赖关系。在意图融合排序模块中，分析会话重复点击和探索行为，并聚合所有级别的意图单元进一步提高模型推荐性能。在Diginetica和Tmall数据集上的实验结果表明，所提模型在平均倒数排名和精确度指标上相比于最优基线模型提高了2.12%和1.27%，具有较好的推荐性能。

Abstract:

Existing session recommendation models based on Graph Neural Network(GNN)mine the potential information between projects by capturing the complex transformation patterns of projects; however, they rarely consider the cross-session information and high-level information in the current session, so they cannot capture the complex dependencies in the session. To solve this problem, a session recommendation model based on global graph and multi-granularity intention unit is built. A cross-session graph is constructed, and a cross-session representation is obtained by using a graph attention network. A multi-granularity intention unit heterogeneous session graph on continuous intention unit is constructed, and the global and local representations are obtained. The cross-session, global, and local representations are fused to capture the complex dependencies between projects in a session. In the intention fusion sorting module, the repeated clicking and exploration behaviors of the session are analyzed, and all levels of intention units are aggregated to further improve the model recommendation performance. Experimental results on the Diginetica and Tmall data sets show that the proposed model is 2.12% and 1.27% higher than the optimal baseline model in terms of Mean Reciprocal Rank(MRR) index and precision index, respectively, and has good recommendation performance.

Key words: recommendation system, session recommendation, Graph Neural Network(GNN), global graph, multi-granularity intention unit

李婉桦, 孙英娟, 刘艺璇, 刘乾. 基于全局图和多粒度意图单元的会话推荐[J]. 计算机工程, 2023, 49(10): 136-144, 153.

Wanhua LI, Yingjuan SUN, Yixuan LIU, Qian LIU. Session Recommendation Based on Global Graph and Multi-granularity Intention Unit[J]. Computer Engineering, 2023, 49(10): 136-144, 153.

http://www.ecice06.com/CN/Y2023/V49/I10/136

图/表 11

图1 GGMIU模型框架

Fig.1 Framework of GGMIU model

图2 同一会话和不同会话之间的项目信息

Fig.2 Project informations between the same session and different sessions

图3 跨会话图的构造

Fig.3 Construction of cross-session graph

图4 连续意图单元、

$ k $

级意图会话图和多粒度意图单元异构会话图的构造

Fig.4 Construction of continuous intention unit, k-level intention session graph and multi-granularity intention unit heterogeneous session graph

图5 Tmall数据集上的实验结果

Fig.5 Experimental results on the Tmall data set

图6 Diginetica数据集上的实验结果

Fig.6 Experimental results on the Diginetica data set

图7 在Tmall和Diginetica数据集上的不同丢弃率分析

Fig.7 Analysis of different dropouts on the Tmall and Diginetica data sets

图8 隐向量维度

$ \boldsymbol{d} $

对模型效果的影响

Fig.8 Influence of hidden vector dimension d on model effect

参考文献 25

1	陈冬梅, 卜霄菲, 黄河, 等. 基于候客点规划的空闲出租车路线推荐算法. 计算机工程, 2022, 48(2): 297- 305. URL
	CHEN D M, BU X F, HUANG H, et al. Idle taxi route recommendation algorithm based on waiting point planning. Computer Engineering, 2022, 48(2): 297- 305. URL
2	周炜翔, 张雯, 杨博, 等. 面向微博用户的个性化推荐算法研究. 计算机工程, 2020, 46(10): 60-66, 73. URL
	ZHOU W X, ZHANG W, YANG B, et al. Research on personalized recommendation algorithm for microblog users. Computer Engineering, 2020, 46(10): 60-66, 73. URL
3	LIAO C L, LEE S J. A clustering based approach to improving the efficiency of collaborative filtering recommendation. Electronic Commerce Research and Applications, 2016, 18, 1- 9. doi: 10.1016/j.elerap.2016.05.001
4	张浩博, 薛峰, 刘凯. 基于半自动编码器的协同过滤推荐算法. 计算机工程, 2021, 47(3): 125- 130. URL
	ZHANG H B, XUE F, LIU K. Collaborative filtering recommendation algorithm based on semi-autoencoder. Computer Engineering, 2021, 47(3): 125- 130. URL
5	SOARES M, VIANA P. Tuning metadata for better movie content-based recommendation systems. Multimedia Tools and Applications, 2015, 74(17): 7015- 7036. doi: 10.1007/s11042-014-1950-1
6	GU Y P, ZHAO B, HARDTKE D, et al. Learning global term weights for content-based recommender systems[C]//Proceedings of the 25th International Conference on World Wide Web. New York, USA: ACM Press, 2016: 391-400.
7	陈恩华, 方宝富. 基于物品嵌入向量的会话型推荐算法. 计算机工程, 2021, 47(7): 74- 80. URL
	CHEN E H, FANG B F. Session-based recommendation algorithm with Item2Vec. Computer Engineering, 2021, 47(7): 74- 80. URL
8	WANG S, CAO L, WANG Y, et al. A survey on session-based recommender systems. ACM Computing Surveys, 2021, 54(7): 1- 38.
9	WANG Z, YE W, CHEN X, et al. Generative session-based recommendation[C]//Proceedings of 2022 ACM Web Conference. New York, USA: ACM Press, 2022: 2227-2235.
10	YANG Y D, PAN X A, YAO X, et al. PHR: a personalized hidden route recommendation system based on hidden Markov model[M]. Berlin, Germany: Springer International Publishing, 2020: 535-539.
11	CHEN J, WANG C K, WANG J M. A personalized interest-forgetting Markov model for recommendations[C]//Proceedings of AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2015: 1-10.
12	WU X, LIU Q, CHEN E H, et al. Personalized next-song recommendation in online Karaokes[C]//Proceedings of the 7th ACM Conference on Recommender Systems. New York, USA: ACM Press, 2013: 137-140.
13	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.
14	HIDASI B, KARATZOGLOU A, BALTRUNAS L, et al. Session-based recommendations with recurrent neural networks[EB/OL]. [2022-07-15]. https://arxiv.org/abs/1511.06939.
15	LI J, REN P J, CHEN Z M, et al. Neural attentive session-based recommendation[C]//Proceedings of 2017 ACM Conference on Information and Knowledge Management. New York, USA: ACM Press, 2017: 1419-1428.
16	LIU Q, ZENG Y F, MOKHOSI R, et al. STAMP: short-term attention/memory priority model for session-based recommendation[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. New York, USA: ACM Press, 2018: 1831-1839.
17	WU S, TANG Y Y, ZHU Y Q, et al. Session-based recommendation with graph neural networks[C]//Proceedings of AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2019: 346-353.
18	QIU R H, LI J J, HUANG Z, et al. Rethinking the item order in session-based recommendation with graph neural networks[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. New York, USA: ACM Press, 2019: 579-588.
19	WANG Z Y, 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, USA: ACM Press, 2020: 169-178.
20	XIA X, YIN H Z, YU J L, et al. Self-supervised hypergraph convolutional networks for session-based recommendation[C]//Proceedings of AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2021: 4503-4511.
21	GUO J Y, YANG Y M, SONG X C, et al. Learning multi-granularity consecutive user intent unit for session-based recommendation[C]//Proceedings of the 15th ACM International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2022: 343-352.
22	WANG Z, WEI W, CONG G, et al. Exploring global information for session-based recommendation[EB/OL]. [2022-07-15]. https://arxiv.org/abs/2011.10173.
23	LI A S. Transition information enhanced disentangled graph neural networks for session-based recommendation. SSRN Electronic Journal, 2022, 210, 118336.
24	REN P J, CHEN Z M, LI J, et al. RepeatNet: a repeat aware neural recommendation machine for session-based recommendation[C]//Proceedings of AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2019: 4806-4813.
25	WANG Z, WEI W, CONG G, et al. Exploiting repeated behavior pattern and long-term item dependency for session-based recommendation[EB/OL]. [2022-07-15]. https://arxiv.org/abs/2012.05422v1.

[1]	隋国华, 李陶然, 刘昊, 陈林, 汪卫. 基于图表示学习的领域知识图谱推理技术研究[J]. 计算机工程, 2023, 49(9): 89-98.
[2]	刘晓黎, 王轶彤. 基于自监督学习的多密度图会话推荐[J]. 计算机工程, 2023, 49(9): 60-68, 78.
[3]	赵世豪, 毛国君, 熊保平, 黄山, 林江宏. 基于图小波卷积神经网络的时空图挖掘模型[J]. 计算机工程, 2023, 49(7): 85-93.
[4]	唐彦, 卢镘旭. 基于知识图谱与深度涟漪网络的推荐系统[J]. 计算机工程, 2023, 49(5): 63-72,80.
[5]	马月坤, 张可心, 高唱. 体现辨证论治差异的不孕症知识图谱构建方法研究[J]. 计算机工程, 2023, 49(3): 280-287,295.
[6]	张晓晖, 马慧芳, 王文涛, 高子皓. 基于跨会话知识图谱的图注意力网络推荐方法[J]. 计算机工程, 2023, 49(2): 136-142,149.
[7]	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.
[8]	雷李想, 武志昊, 刘钰, 周子站. 基于域内特征间相似性的点击率预估优化[J]. 计算机工程, 2023, 49(2): 238-245.
[9]	潘嘉诚, 董一鸿, 陈华辉. 基于图神经网络的自闭症辅助诊断研究综述[J]. 计算机工程, 2022, 48(9): 1-11.
[10]	丁庆丰, 李晋国. 一种物联网环境下的分布式异常流量检测方案[J]. 计算机工程, 2022, 48(8): 152-159.
[11]	胡承佐, 王庆梅, 李迪超, 王铮. 基于复杂结构信息的图神经网络序列推荐算法[J]. 计算机工程, 2022, 48(5): 82-90,97.
[12]	金雨澄, 王清钦, 高剑, 苗仲辰, 林越峰, 项雅丽, 熊贇. 基于图深度学习的金融文本多标签分类算法[J]. 计算机工程, 2022, 48(4): 16-21.
[13]	赵越, 武志昊, 赵苡积. 基于特征与域感知的点击率预估方法[J]. 计算机工程, 2022, 48(3): 60-68.
[14]	崔丽平, 古丽拉·阿东别克, 王智悦. 基于有向图模型的旅游领域命名实体识别[J]. 计算机工程, 2022, 48(2): 306-313.
[15]	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[J]. 计算机工程, 2022, 48(11): 55-61.

选择文件类型/文献管理软件名称

选择包含的内容