| 摘要: |
| 共享账户是指多个用户通常共享同一电子账户(智能家居系统往往以一个家庭为一个账户), 因此, 同一账户下的行为序列往往包含多个用户的交互记录. 面向共享账户场景的序列推荐成为近年来的研究热点, 但仍然面临以下挑战: (1) 现有方法难以有效解耦同一账户下的不同潜在用户偏好, 导致用户表示学习出现偏差. (2) 现有方法通常采用固定的超参数来模拟共享账户中的潜在用户数量, 难以自适应地确定潜在用户数量, 易导致欠拟合或噪声干扰问题. 为了解决上述挑战, 提出一种面向共享账户序列推荐的提示增强图注意力网络(PE-GAT). 具体而言, 该方法首先通过设计动态加权密度聚类算法来确定账户中的潜在用户数量, 构建用户级序列图, 以显式区分不同潜在用户的交互行为. 在此基础上, 提出基于图注意力网络的预训练模型, 通过解耦混合序列中的用户偏好来学习初始序列表示. 随后, 受预训练-提示调优范式启发, 设计了基于注意力机制的提示增强模块, 通过提示模板优化序列表示并指导模型再训练. 最终, PE-GAT通过融合账户级表示与提示增强后的序列表示, 以生成个性化的推荐结果. 在HVIDEO-E和HVIDEO-V两个真实共享账户数据集上的实验表明, PE-GAT优于14种基线方法, 在MRR和Recall指标上分别最高提升4.73%和5.59%. |
| 关键词: 推荐系统 共享账户序列推荐 提示增强 图注意力网络 |
| DOI：10.13328/j.cnki.jos.007626 |
| 分类号:TP18 |
| 基金项目:国家自然科学基金(62472263, 52374221, 62502288); 山东省泰山学者人才工程(tstp20250506, tsqn202211154); 山东省自然科学基金项目(ZR2024MF034, ZR2022MF268) |
|
| Prompt-enhanced Graph Attention Network for Shared-account Sequential Recommendation |
|
ZHAO Zhong-Ying1, WANG Ying1, ZHANG Jin-Yu1, ZHOU Hui1, LI Chao2, ZENG Qing-Tian1
|
|
1.College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China;2.College of Electronic and Information Engineering, Shandong University of Science and Technology, Qingdao 266590, China
|
| Abstract: |
| In real-world scenarios such as e-commerce platforms and smart-home systems, multiple users often share the same account, resulting in behavioral sequences that mix interactions from different users. Sequential recommendation in shared-account settings has recently attracted increasing research attention but still faces two major challenges: (1) interactions from different latent users are difficult to disentangle accurately, leading to biased representation learning; (2) reliance on static hyperparameters to estimate the number of latent users prevents adaptive determination of user cardinality, which may cause underfitting or introduce noise. To address these challenges, this study proposes PE-GAT (prompt-enhanced graph attention network), an adaptive sequential recommendation framework for shared-account settings. PE-GAT first employs a dynamically-weighted, density-based clustering algorithm to infer the number of latent users and construct user-level sequential graphs that explicitly separate heterogeneous user behaviors. Based on these graphs, a graph attention network pre-training module is employed to disentangle mixed user preferences and learn initial sequence representations. Inspired by the pre-training and prompt-tuning paradigm, a self-attention-based prompt-enhancement module is then designed to refine sequence embeddings via prompt templates during retraining. Finally, account-level representations are fused with prompt-enhanced sequence embeddings to generate personalized recommendations. Extensive experiments on two real-world shared-account datasets (HVIDEO-E and HVIDEO-V) show that PE-GAT outperforms 14 state-of-the-art baselines, achieving maximum improvements of 4.73% and 5.59% in MRR and Recall metrics, respectively. |
| Key words: recommender system shared-account sequential recommendation (SSR) prompt enhanced graph attention network (GAT) |
