Reinforcement Learning-enhanced Shared-account Cross-domain Sequential Recommendation

Abstract

Shared-account Cross-domain Sequential Recommendation (SCSR) is an emerging yet challenging task that simultaneously considers the shared-account and cross-domain characteristics in the sequential recommendation. Existing works on SCSR are mainly based on Recurrent Neural Network (RNN) and Graph Neural Network (GNN) but they ignore the fact that although multiple users share a single account, it is mainly occupied by one user at a time. This observation motivates us to learn a more accurate user-specific account representation by attentively focusing on its recent behaviors. Furthermore, though existing works endow lower weights to irrelevant interactions, they may still dilute the domain information and impede the cross-domain recommendation. To address the above issues, we propose a reinforcement learning-based solution, namely RL-ISN, which consists of a basic cross-domain recommender and a reinforcement learning-based domain filter. Specifically, to model the account representation in the shared-account scenario, the basic recommender first clusters users' mixed behaviors as latent users, and then leverages an attention model over them to conduct user identification. To reduce the impact of irrelevant domain information, we formulate the domain filter as a hierarchical reinforcement learning task, where a high-level task is utilized to decide whether to revise the whole transferred sequence or not, and if it does, a low-level task is further performed to determine whether to remove each interaction within it or not. To evaluate the performance of our solution, we conduct extensive experiments on two real-world datasets, and the experimental results demonstrate the superiority of our RL-ISN method compared with the state-of-the-art recommendation methods.

O‐SCDR: Optimal cluster with attention based shared‐account cross‐domain sequential recommendation using deep reinforcement learning technique

“…It outperforms NAIS (He et al, 2018) by 70.535%, π‐Net (Ma et al, 2019) by 66.45% and PSJ‐net (Sun et al, 2021) by 68.06%. Moreover, it surpasses DA‐GCN (Guo et al, 2021) by 70.925%, ISN‐RL (Guo, Zhang, Chen, et al, 2022) by 66.165%, and even RL‐ISN (Guo, Zhang, Chen, et al, 2022) by 64.11%. Impressively, the culmination of these results is seen in the comparison of our QCDRL method with all the evaluated techniques at @25 recommendations, where it showcases an unparalleled efficiency improvement of 47.99% over the best‐performing existing technique, RL‐ISN (Guo, Zhang, Chen, et al, 2022).…”

“…For sequential recommendation, the methods assessed are GRU4REC (Hidasi et al, 2015), HGRU4REC (Quadrana et al, 2017), and NAIS (He et al, 2018). Additionally, the SCSR methods evaluated include π‐Net (Ma et al, 2019), PSJ‐net (Sun et al, 2021), DA‐GCN (Guo et al, 2021), ISN‐RL (Guo, Zhang, Chen, et al, 2022), and RL‐ISN (Guo, Zhang, Chen, et al, 2022). The evaluation employs the widely recognized hit ratio and normalized discounted cumulative gain as the key metrics.…”

“…During this process, users pursue specific objectives, such as discovering a desirable restaurant in a city or exploring music of a particular genre or mood. Guo, Zhang, Chen, et al (2022) introduced a reinforcement learning‐based approach labelled RL‐ISN. This innovative solution involves a foundational SCSR combined with a reinforcement learning‐driven domain filter (Zeyu et al, 2023).…”

“…Second, they heavily rely on explicit user ratings, which are frequently absent in SCSR scenarios. Recent works such as TiDA‐GCN (Guo, Zhang, Tang, et al, 2022) and RL‐ISN (Guo, Zhang, Chen, et al, 2022) have emerged, specifically tailored for SCSR. However, their dependence on neural networks may limit their ability to adequately capture complex associations or model the structural information necessary for bridging the gap between two domains.…”

Self‐supervised graph learning for occasional group recommendation

“…The goal of group recommendation is to recommend proper items to a group. Different from Shared-account recommendation, 45,46 where the members in the shared-account is closely related, the members in the group may formed at hoc. Existing methods for group recommendation can be classified into the following two categories: 20 and (B, D, F…”