Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

NFARec: A Negative Feedback-Aware Recommender Model

Xinfeng Wang, Fumiyo Fukumoto, Jin Cui, Yoshimi Suzuki, Dongjin Yu

TL;DR

This paper proposes a negative feedback-aware recommender model (NFARec) that maximizes the leverage of negative feedback and adopts a feedback-aware correlation that guides hypergraph convolutions (HGCs) to learn users' structural representations.

Abstract

Graph neural network (GNN)-based models have been extensively studied for recommendations, as they can extract high-order collaborative signals accurately which is required for high-quality recommender systems. However, they neglect the valuable information gained through negative feedback in two aspects: (1) different users might hold opposite feedback on the same item, which hampers optimal information propagation in GNNs, and (2) even when an item vastly deviates from users' preferences, they might still choose it and provide a negative rating. In this paper, we propose a negative feedback-aware recommender model (NFARec) that maximizes the leverage of negative feedback. To transfer information to multi-hop neighbors along an optimal path effectively, NFARec adopts a feedback-aware correlation that guides hypergraph convolutions (HGCs) to learn users' structural representations. Moreover, NFARec incorporates an auxiliary task - predicting the feedback sentiment polarity (i.e., positive or negative) of the next interaction - based on the Transformer Hawkes Process. The task is beneficial for understanding users by learning the sentiment expressed in their previous sequential feedback patterns and predicting future interactions. Extensive experiments demonstrate that NFARec outperforms competitive baselines. Our source code and data are released at https://github.com/WangXFng/NFARec.

NFARec: A Negative Feedback-Aware Recommender Model

TL;DR

This paper proposes a negative feedback-aware recommender model (NFARec) that maximizes the leverage of negative feedback and adopts a feedback-aware correlation that guides hypergraph convolutions (HGCs) to learn users' structural representations.

Abstract

Graph neural network (GNN)-based models have been extensively studied for recommendations, as they can extract high-order collaborative signals accurately which is required for high-quality recommender systems. However, they neglect the valuable information gained through negative feedback in two aspects: (1) different users might hold opposite feedback on the same item, which hampers optimal information propagation in GNNs, and (2) even when an item vastly deviates from users' preferences, they might still choose it and provide a negative rating. In this paper, we propose a negative feedback-aware recommender model (NFARec) that maximizes the leverage of negative feedback. To transfer information to multi-hop neighbors along an optimal path effectively, NFARec adopts a feedback-aware correlation that guides hypergraph convolutions (HGCs) to learn users' structural representations. Moreover, NFARec incorporates an auxiliary task - predicting the feedback sentiment polarity (i.e., positive or negative) of the next interaction - based on the Transformer Hawkes Process. The task is beneficial for understanding users by learning the sentiment expressed in their previous sequential feedback patterns and predicting future interactions. Extensive experiments demonstrate that NFARec outperforms competitive baselines. Our source code and data are released at https://github.com/WangXFng/NFARec.
Paper Structure (19 sections, 16 equations, 5 figures, 5 tables)

This paper contains 19 sections, 16 equations, 5 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Preliminaries
  4. Definitions
  5. Hawkes Process
  6. Enhanced HGC with Feedback
  7. Methodology
  8. Sequential Representation Learning
  9. Structural Representation Learning
  10. Sequential and Structural Decoder
  11. Model Optimization
  12. Experiment
  13. Experimental Setup
  14. Overall Performance
  15. Ablation Study
  16. ...and 4 more sections

Figures (5)

  • Figure 2: Architecture of our NFARec framework. NFARec learns negative feedback in both sequential and structural patterns.
  • Figure 3: Illustration of various feedback.
  • Figure 4: Visualizations of users’ sequential and structure representations (Reps.) and item representations.
  • Figure 5: Effect of various $\delta$ on performance.
  • Figure 6: Case Study. ✔ and ✘ indicate that the model recommends correctly and incorrectly, respectively. ✘ denotes that the removed component is the key in each case. We do not show the case study on Gra$_1$, as it is not our main focus.