TransRec++: Translation-based sequential recommendation with heterogeneous feedback

doi:10.1007/s11704-022-1184-8

Front. Comput. Sci.

2022, Vol. 16

Issue (2) : 162615 https://doi.org/10.1007/s11704-022-1184-8

LETTER

TransRec++: Translation-based sequential recommendation with heterogeneous feedback

Zhuo-Xin ZHAN^1,², Ming-Kai HE^1,², Wei-Ke PAN^1,²(

), Zhong MING^1,²

¹. National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, Shenzhen 518060, China
². College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China

Download: PDF(1048 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Corresponding Author(s): Wei-Ke PAN

Just Accepted Date: 29 October 2021 Issue Date: 01 March 2022

Cite this article:

Zhuo-Xin ZHAN,Ming-Kai HE,Wei-Ke PAN, et al. TransRec++: Translation-based sequential recommendation with heterogeneous feedback[J]. Front. Comput. Sci., 2022, 16(2): 162615.

URL:

https://academic.hep.com.cn/fcs/EN/10.1007/s11704-022-1184-8
https://academic.hep.com.cn/fcs/EN/Y2022/V16/I2/162615

Fig.1 Illustration of our TransRec++, which first embeds items into vectors in a transition space and then learns user-specific behavior transition vectors with different orders

1 ≤ ? ≤ L

Fig.2

Tab.1 Main results on UB and Tmall

$L$	$# b h$	UB		Tmall
$L$	$# b h$	HR@10	NDCG@10	HR@10	NDCG@10
1	2	0.0661	0.0413	0.0540	0.0387
1	4	0.0729	0.0466	0.0546	0.0401
2	2	0.0647	0.0390	0.0593	0.0377
2	4	0.0760	0.0451	0.0614	0.0385
3	2	0.0627	0.0355	0.0545	0.0324
3	4	0.0726	0.0429	0.0570	0.0337

Tab.2 Recommendation performance with more types of behaviors

Fig.3 Recommendation performance of deep learning-based methods (i.e., RIB and BINN, denoted by “origin”) and their enhanced versions with behavior transition vectors (i.e., RIB++ and BINN++, denoted by “origin++”)

1	R N He, W C Kang, J McAuley. Translation-based recommendation. In: Proceedings of the 17th ACM Conference on Recommender Systems. 2017, 161– 169
2	X G Peng, Y F Chen, Y C Duan, W K Pan, Z Ming. RBPR: role-based Bayesian personalized ranking for heterogeneous one-class collaborative filtering. In: Proceedings of the 24th ACM Conference on User Modeling, Adaptation and Personalisation (Extended Proceedings). 2016
3	W K Pan , Q Yang , W L Cai , Y F Chen , Q Zhang , X G Peng , Z Ming . Transfer to rank for heterogeneous one-class collaborative filtering. ACM Transactions on Information Systems, 2019, 37( 1): 10–
4	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
5	R N He, J McAuley. Fusing similarity models with Markov chains for sparse sequential recommendation. In: Proceedings of 16th International Conference on Data Mining. 2016, 191– 200
6	M Z Zhou, Z Y Ding, J L Tang, D W Yin. Micro behaviors: a new perspective in e-commerce recommender systems. In: Proceedings of the 11th ACM International Conference on Web Search and Data Mining. 2018, 727– 735
7	Z Li, H K Zhao, Q Liu, Z Y Huang, T Mei, E H Chen. Learning from history and present: Next-item recommendation via discriminatively exploiting user behaviors. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018, 1734−1743

[1]		Download
[2]		Download

Viewed

Full text

Abstract

Cited

Shared

Discussed