CPGRec+:兼顾准确性与多样性的个性化视频游戏推荐框架
CPGRec+: A Balance-oriented Framework for Personalized Video Game Recommendations
CPGRec+ 是一个面向视频游戏推荐的平衡导向框架,针对既有 GNN 方法偏重准确性、忽视多样性以及过平滑问题提出改进。其新增 PER 模块通过带符号边权重区分玩家兴趣与不兴趣并量化偏好强度,PRG 模块利用 LLM 推理生成游戏与玩家的上下文描述以优化表示。在两个 Steam 数据集上的实验显示,其准确性和多样性均超过 SOTA 模型,代码已开源。
Abstract:The rapid expansion of gaming industry requires advanced recommender systems tailored to its dynamic landscape. Existing Graph Neural Network (GNN)-based methods primarily prioritize accuracy over diversity, overlooking their inherent trade-off. To address this, we previously proposed CPGRec, a balance-oriented gaming recommender system. However, CPGRec fails to account for critical disparities in player-game interactions, which carry varying significance in reflecting players' personal preferences and may exacerbate over-smoothness issues inherent in GNN-based models. Moreover, existing approaches underutilize the reasoning capabilities and extensive knowledge of large language models (LLMs) in addressing these limitations. To bridge this gap, we propose two new modules. First, Preference-informed Edge Reweighting (PER) module assigns signed edge weights to qualitatively distinguish significant player interests and disinterests while then quantitatively measuring preference strength to mitigate over-smoothing in graph convolutions. Second, Preference-informed Representation Generation (PRG) module leverages LLMs to generate contextualized descriptions of games and players by reasoning personal preferences from comparing global and personal interests, thereby refining representations of players and games. Experiments on \textcolor{black}{two Steam datasets} demonstrate CPGRec+'s superior accuracy and diversity over state-of-the-art models. The code is accessible at this https URL.
| Comments: | Published in ACM Transactions on Information Systems (TOIS). 43 pages, 9 figures. Added references on multi-view clustering |
| Subjects: | Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2604.14586 [cs.IR] |
| (or arXiv:2604.14586v5 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2604.14586 arXiv-issued DOI via DataCite |
|
| Journal reference: | ACM Trans. Inf. Syst. 44, 3, Article 66 (March 2026), 44 pages |
| Related DOI: | https://doi.org/10.1145/378926
DOI(s) linking to related resources |
Submission history
From: Xiping Li [view email]
[v1]
Thu, 16 Apr 2026 03:25:18 UTC (6,130 KB)
[v2]
Sun, 19 Apr 2026 01:21:34 UTC (6,131 KB)
[v3]
Tue, 21 Jul 2026 17:16:37 UTC (5,422 KB)
[v4]
Thu, 27 Aug 2026 10:13:06 UTC (5,422 KB)
[v5]
Fri, 13 Nov 2026 15:35:46 UTC (5,408 KB)
来源:arXiv cs.AI · arxiv.org