Researchers have developed GSPRec, a novel framework for collaborative filtering that enhances item representations by incorporating item-item proximity derived from user interaction sequences. Unlike previous methods that treat collaborative filtering as a low-pass filter, GSPRec utilizes intermediate-frequency components to capture community-level user preferences. The framework constructs a unified graph topology by integrating user-item interactions with item-item edges derived from ordered user interactions, which are then strengthened through multi-hop diffusion. Experiments on four real-world datasets demonstrated that GSPRec outperforms existing graph-based collaborative filtering baselines, achieving an average improvement of 5.12% in NDCG@10. AI
IMPACT This research could lead to more accurate recommendation systems by improving how item relationships are modeled.
RANK_REASON The cluster contains an academic paper detailing a new framework and experimental results. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
