BA-GNN: Behavior-aware graph neural network for session-based recommendation

doi:10.1007/s11704-022-2324-x

Front. Comput. Sci.

2023, Vol. 17

Issue (6) : 176613 https://doi.org/10.1007/s11704-022-2324-x

Information Systems

BA-GNN: Behavior-aware graph neural network for session-based recommendation

Yongquan LIANG¹, Qiuyu SONG¹, Zhongying ZHAO¹(

), Hui ZHOU², Maoguo GONG^1,²(

)

¹. School of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China
². School of Electronic Engineering, Xidian University, Xi’an 710071, China

Download: PDF(10235 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Session-based recommendation is a popular research topic that aims to predict users’ next possible interactive item by exploiting anonymous sessions. The existing studies mainly focus on making predictions by considering users’ single interactive behavior. Some recent efforts have been made to exploit multiple interactive behaviors, but they generally ignore the influences of different interactive behaviors and the noise in interactive sequences. To address these problems, we propose a behavior-aware graph neural network for session-based recommendation. First, different interactive sequences are modeled as directed graphs. Thus, the item representations are learned via graph neural networks. Then, a sparse self-attention module is designed to remove the noise in behavior sequences. Finally, the representations of different behavior sequences are aggregated with the gating mechanism to obtain the session representations. Experimental results on two public datasets show that our proposed method outperforms all competitive baselines. The source code is available at the website of GitHub.

Keywords session-based recommendation multiple interactive behaviors graph neural networks

Corresponding Author(s): Zhongying ZHAO,Maoguo GONG

Just Accepted Date: 09 October 2022 Issue Date: 13 February 2023

Cite this article:

Yongquan LIANG,Qiuyu SONG,Zhongying ZHAO, et al. BA-GNN: Behavior-aware graph neural network for session-based recommendation[J]. Front. Comput. Sci., 2023, 17(6): 176613.

URL:

https://academic.hep.com.cn/fcs/EN/10.1007/s11704-022-2324-x
https://academic.hep.com.cn/fcs/EN/Y2023/V17/I6/176613

Fig.1 A toy sample of users’ session interactions

Notations	Descriptions
$V$	The item set.
$m$	The number of items.
$s$	The sequence of items for session $s$ .
$s (c)$	The click sequence of session $s$ .
$\| s (c) \|$	The number of items in the click sequence $s (c)$ .
$s (p)$	The purchase sequence of session $s$ .
$\| s (p) \|$	The number of items in the purchase sequence $s (p)$ .
$G c$	The graph of the click sequence $s (c)$ .
$G p$	The graph of the purchase sequence $s (p)$ .
$A c$	The adjacency matrix of graph $G c$ .
$A p$	The adjacency matrix of graph $G p$ .
$v c, i$	The representation of node $v c, i$ .
$v p, i$	The representation of node $v p, i$ .
$v g, i$	The global representation of node $v i$ .
$z s (c)$	The representation of the click sequence $s (c)$ .
$z s (p)$	The representation of the purchase sequence $s (p)$ .
$α$	The weight of the click sequence $s (c)$ .
$s$	The representation of the session $s$ .

Tab.1 The notations used in this paper

Fig.2 The overall framework of our proposed method. We first construct directed graphs for different action sequences. Then we learn the behavior-level item representations via GGNN. Next, a sparse self-attention module is designed to remove the noise in the click sequences. After that, we learn representations for action sequences and sessions. Finally, we make predictions and recommendations

Tab.2 Statistics of the datasets

Tab.3 Results of all methods

Tab.4 Evaluation results for different top-K recommendations

Tab.5 Ablation experiment

Fig.3 The effect of the maximum sequence length L

Fig.4 The effect of GGNN propagation step t

1	S, Wang L, Cao Y, Wang Q Z, Sheng M A, Orgun D Lian . A survey on session-based recommender systems. ACM Computing Surveys, 2022, 54( 7): 154:1-154:38
2	B, Hidasi A, Karatzoglou L, Baltrunas D Tikk . Session-based recommendations with recurrent neural networks. In: Proceedings of the 4th International Conference on Learning Representations. 2016, 1–10
3	Li J, Ren P, Chen Z, Ren Z, Lian T, Ma J. Neural attentive session-based recommendation. In: Proceedings of 2017 ACM Conference on Information and Knowledge Management. 2017, 1419–1428
4	Z, Wang W, Wei G, Cong X L, Li X L, Mao M Qiu . Global context enhanced graph neural networks for session-based recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2020, 169–178
5	S, Wu Y, Tang Y, Zhu L, Wang X, Xie T Tan . Session-based recommendation with graph neural networks. In: Proceedings of the 33rd AAAI Conference on Artificial Intelligence. 2019, 346–353
6	C, Xu P, Zhao Y, Liu V S, Sheng J, Xu F, Zhuang J, Fang X Zhou . Graph contextualized self-attention network for session-based recommendation. In: Proceedings of the 28th International Joint Conference on Artificial Intelligence. 2019, 3940–3946
7	F, Yu Y, Zhu Q, Liu S, Wu L, Wang T Tan . TAGNN: target attentive graph neural networks for session-based recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2020, 1921–1924
8	W, Wang W, Zhang S, Liu Q, Liu B, Zhang L, Lin H Zha . Beyond clicks: modeling multi-relational item graph for session-based target behavior prediction. In: Proceedings of the Web Conference 2020. 2020, 3056–3062
9	G, Shani D, Heckerman R I. Brafman . An MDP-based recommender system. Journal of Machine Learning Research, 2005, 6(9): 1265–1295
10	S, Rendle C, Freudenthaler L Schmidt-Thieme . Factorizing personalized Markov chains for next-basket recommendation. In: Proceedings of the 19th International Conference on World Wide Web. 2010, 811–820
11	D, Garg P, Gupta P, Malhotra L, Vig G Shroff . Sequence and time aware neighborhood for session-based recommendations: STAN. In: Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2019, 1069–1072
12	D, Jannach M Ludewig . When recurrent neural networks meet the neighborhood for session-based recommendation. In: Proceedings of the 11th ACM Conference on Recommender Systems. 2017, 306–310
13	Y K, Tan X, Xu Y Liu . Improved recurrent neural networks for session-based recommendations. In: Proceedings of the 1st Workshop on Deep Learning for Recommender Systems. 2016, 17–22
14	Q, Liu Y, Zeng R, Mokhosi H Zhang . STAMP: short-term attention/memory priority model for session-based recommendation. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018, 1831–1839
15	J, Song H, Shen Z, Ou J, Zhang T, Xiao S Liang . ISLF: interest shift and latent factors combination model for session-based recommendation. In: Proceedings of the 28th International Joint Conference on Artificial Intelligence. 2019, 5765–5771
16	S, Wang L, Hu Y, Wang Q Z, Sheng M, Orgun L Cao . Modeling multi-purpose sessions for next-item recommendations via mixture-channel purpose routing networks. In: Proceedings of the 28th International Joint Conference on Artificial Intelligence. 2019, 3771–3777
17	R, Qiu J, Li Z, Huang H Yin . Rethinking the item order in session-based recommendation with graph neural networks. In: Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 2019, 579–588
18	T R, Gwadabe Y Liu . Improving graph neural network for session-based recommendation system via non-sequential interactions. Neurocomputing, 2022, 468: 111–122
19	D T, Le H W, Lauw Y Fang . Modeling contemporaneous basket sequences with twin networks for next-item recommendation. In: Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018, 3414–3420
20	W, Meng D, Yang Y Xiao . Incorporating user micro-behaviors and item knowledge into multi-task learning for session-based recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2020, 1091–1100
21	Y, Li D, Tarlow M, Brockschmidt R S Zemel . Gated graph sequence neural networks. In: Proceedings of the 4th International Conference on Learning Representations. 2016, 1–20
22	Bridle J S. Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition. Neurocomputing, 1990, 68 : 227–236
23	A F T, Martins R F Astudillo . From Softmax to Sparsemax: a sparse model of attention and multi-label classification. In: Proceedings of the 33rd International Conference on Machine Learning. 2016, 1614–1623
24	J, Yuan Z, Song M, Sun X, Wang W X Zhao . Dual sparse attention network for session-based recommendation. In: Proceedings of the 35th AAAI Conference on Artificial Intelligence. 2021, 4635–4643
25	A, Vaswani N, Shazeer N, Parmar J, Uszkoreit L, Jones A N, Gomez Ł, Kaiser I Polosukhin . Attention is all you need. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, 6000–6010
26	B, Sarwar G, Karypis J, Konstan J Riedl . Item-based collaborative filtering recommendation algorithms. In: Proceedings of the 10th International Conference on World Wide Web. 2001, 285–295

[1]

FCS-22324-OF-JL_suppl_1

Download

[1]	Jinwei LUO, Mingkai HE, Weike PAN, Zhong MING. BGNN: Behavior-aware graph neural network for heterogeneous session-based recommendation[J]. Front. Comput. Sci., 2023, 17(5): 175336-.
[2]	Yuan GAO, Xiang WANG, Xiangnan HE, Huamin FENG, Yongdong ZHANG. Rumor detection with self-supervised learning on texts and social graph[J]. Front. Comput. Sci., 2023, 17(4): 174611-.

Viewed

Full text

Abstract

Cited

Shared

Discussed