New framework CPGRec+ enhances game recommendations using LLMs and GNNs

Researchers have developed CPGRec+, an enhanced framework for personalized video game recommendations that addresses limitations in existing Graph Neural Network (GNN) models. This new framework incorporates a Preference-informed Edge Reweighting (PER) module to better distinguish significant player interactions and a Preference-informed Representation Generation (PRG) module that utilizes large language models (LLMs) to generate contextualized descriptions. Experiments conducted on Steam datasets indicate that CPGRec+ outperforms state-of-the-art models in both accuracy and diversity of recommendations. AI

IMPACT This framework could improve personalized content delivery in gaming and other domains by better balancing accuracy and diversity.

RANK_REASON The cluster contains a research paper detailing a new framework for recommender systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CPGRec+
• graph neural network
• large-language models
• Preference-informed Edge Reweighting
• Preference-informed Representation Generation
• Steam

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →