A Unified Model for Recommendation with Selective Neighborhood Modeling

TL;DR

This paper introduces a unified neighborhood-based recommendation model that intelligently filters and aggregates neighbor information using a hybrid gated network, improving recommendation accuracy over existing methods.

Contribution

The paper proposes a novel hybrid gated network and user-neighbor regularization to enhance neighborhood modeling in collaborative filtering.

Findings

Consistently outperforms state-of-the-art neighborhood recommenders on three datasets.

Effectively separates similar and dissimilar neighbors to reduce noise impact.

Validates the effectiveness of the hybrid gated network and user-neighbor components.

Abstract

Neighborhood-based recommenders are a major class of Collaborative Filtering (CF) models. The intuition is to exploit neighbors with similar preferences for bridging unseen user-item pairs and alleviating data sparseness. Many existing works propose neural attention networks to aggregate neighbors and place higher weights on specific subsets of users for recommendation. However, the neighborhood information is not necessarily always informative, and the noises in the neighborhood can negatively affect the model performance. To address this issue, we propose a novel neighborhood-based recommender, where a hybrid gated network is designed to automatically separate similar neighbors from dissimilar (noisy) ones, and aggregate those similar neighbors to comprise neighborhood representations. The confidence in the neighborhood is also addressed by putting higher weights on the neighborhood representations if we are confident with the neighborhood information, and vice versa. In addition, a user-neighbor component is proposed to explicitly regularize user-neighbor proximity in the latent space. These two components are combined into a unified model to complement each other for the recommendation task. Extensive experiments on three publicly available datasets show that the proposed model consistently outperforms state-of-the-art neighborhood-based recommenders. We also study different variants of the proposed model to justify the underlying intuition of the proposed hybrid gated network and user-neighbor modeling components.