A Simple yet Effective Negative Sampling Plugin for Constructing Positive Sample Pairs in Implicit Collaborative Filtering

TL;DR

This paper introduces PSP-NS, a simple negative sampling plugin that enhances positive sample construction in implicit collaborative filtering, improving ranking metrics by leveraging interaction confidence and activity-aware weighting.

Contribution

It proposes a novel negative sampling plugin that constructs positive pairs using reweighting and activity-aware schemes, addressing user activity bias and enhancing recommendation quality.

Findings

Boosts Recall@30 by 32.11% on Yelp

Improves Precision@30 by 22.90% on Yelp

Theoretically explains ranking improvements via margin analysis

Abstract

Most implicit collaborative filtering (CF) models are trained with negative sampling, where existing work designs sophisticated strategies for high-quality negatives while largely overlooking the exploration of positive samples. Although some denoising recommendation methods can be applied to implicit CF for denoising positive samples, they often sparsify positive supervision. Moreover, these approaches generally overlook user activity bias during training, leading to insufficient learning for inactive users. To address these issues, we propose a simple yet effective negative sampling plugin, PSP-NS, from the perspective of enhancing positive supervision signals. It builds a user-item bipartite graph with edge weights indicating interaction confidence inferred from global and local patterns, generates positive sample pairs via replication-based reweighting to strengthen positive signals, and adopts an activity-aware weighting scheme to effectively learn inactive users' preferences. We provide theoretical insights from a margin-improvement perspective, explaining why PSP-NS tends to improve ranking quality (e.g., Precision@k/Recall@k), and conduct extensive experiments on four real-world datasets to demonstrate its superiority. For instance, PSP-NS boosts Recall@30 and Precision@30 by 32.11% and 22.90% on Yelp over the strongest baselines. PSP-NS can be integrated with various implicit CF recommenders or negative sampling methods to enhance their performance.