484
Views
17
Downloads
0
Crossref
N/A
WoS
0
Scopus
N/A
CSCD
Session-based recommender systems are increasingly applied to next-item recommendations. However, existing approaches encode the session information of each user independently and do not consider the interrelationship between users. This work is based on the intuition that dynamic groups of like-minded users exist over time. By considering the impact of latent user groups, we can learn a user’s preference in a better way. To this end, we propose a recommendation model based on learning user embeddings by modeling long and short-term dynamic latent user groups. Specifically, we utilize two network units to learn users’ long and short-term sessions, respectively. Meanwhile, we employ two additional units to determine the affiliation of users with specific latent groups, followed by an aggregation of these latent group representations. Finally, user preference representations are shaped comprehensively by considering all these four aspects, based on an attention mechanism. Moreover, to avoid setting the number of groups manually, we further incorporate an adaptive learning unit to assess the necessity for creating a new group and learn the representation of emerging groups automatically. Extensive experiments prove our model outperforms multiple state-of-the-art methods in terms of Recall, mean average precision (mAP), and area under curve (AUC) metrics.
Session-based recommender systems are increasingly applied to next-item recommendations. However, existing approaches encode the session information of each user independently and do not consider the interrelationship between users. This work is based on the intuition that dynamic groups of like-minded users exist over time. By considering the impact of latent user groups, we can learn a user’s preference in a better way. To this end, we propose a recommendation model based on learning user embeddings by modeling long and short-term dynamic latent user groups. Specifically, we utilize two network units to learn users’ long and short-term sessions, respectively. Meanwhile, we employ two additional units to determine the affiliation of users with specific latent groups, followed by an aggregation of these latent group representations. Finally, user preference representations are shaped comprehensively by considering all these four aspects, based on an attention mechanism. Moreover, to avoid setting the number of groups manually, we further incorporate an adaptive learning unit to assess the necessity for creating a new group and learn the representation of emerging groups automatically. Extensive experiments prove our model outperforms multiple state-of-the-art methods in terms of Recall, mean average precision (mAP), and area under curve (AUC) metrics.
X. Zhang, H. Lin, B. Xu, C. Li, Y. Lin, H. Liu, and F. Ma, Dynamic intent-aware iterative denoising network for session-based recommendation, Inf. Process. Manag., vol. 59, no. 3, p. 102936, 2022.
N. Zhu, J. Cao, X. Lu, and H. Xiong, Learning a hierarchical intent model for next-item recommendation, ACM Trans. Inf. Syst., vol. 40, no. 2, pp. 1–28, 2021.
R. Zhang and Y. Mao, Movie recommendation via Markovian factorization of matrix processes, IEEE Access, vol. 7, pp. 13189–13199, 2019.
X. Zhang, H. Lin, L. Yang, B. Xu, Y. Diao, and L. Ren, Dual part-pooling attentive networks for session-based recommendation, Neurocomputing, vol. 440, pp. 89–100, 2021.
W. Chen, P. Ren, F. Cai, F. Sun, and M. D. Rijke, Multi-interest diversification for end-to-end sequential recommendation, ACM Trans. Inf. Syst., vol. 40, no. 1, pp. 1–30, 2021.
This work was partially supported by the National Natural Science Foundation of China (No. 62202282) and Shanghai Youth Science and Technology Talents Sailing Program (No. 22YF1413700).
The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).