Contrastive Learning for Aspect Representation towards Explainable Recommendation

2026-10-06

AI Summary

This study addresses the challenge in existing recommender systems of effectively integrating review semantics with rating information while balancing accuracy and interpretability. To this end, we propose CLARER, a novel framework that employs a multilayer perceptron to process rating features and a Transformer encoder to extract aspect-level features from textual reviews. Crucially, CLARER introduces contrastive learning to enhance the discriminative power of aspect representations, thereby facilitating deep multimodal feature fusion. A decoder subsequently generates natural language explanations for recommendations. Extensive experiments conducted on three benchmark datasets demonstrate that CLARER significantly outperforms state-of-the-art baseline methods in both recommendation accuracy and explanation generation quality, establishing it as an effective approach for interpretable recommendation.

Abstract

In this work, we propose a novel recommendation model, CLARER (Contrastive Learning for Aspect Representation towards Explainable Recommendation) that integrates aspect features learned from textual reviews with rating information to improve the accuracy and explainability of recommendations. Our proposed framework learns user and item representations by combining rating-based features and aspect-based features from reviews. Specifically, rating-based features are learned through a multi-layer perceptron (MLP) model, while aspect-specific review representations are learned using a transformer encoder to capture the semantic information and contrastive learning to better distinguish user preferences. To provide explanations, we train a transformer decoder, using the final representations of users and items from both rating and aspect-based features as context. Experimental results in three benchmark data sets demonstrate that our model achieves superior performance compared to baseline methods in both recommendation (accuracy) and explanation generation.

Problem

• Explainable Recommendation
• Contrastive Learning
• Aspect Representation
• Textual Reviews
• Recommendation Accuracy

Innovation

• Contrastive Learning
• Explainable Recommendation
• Aspect Representation
• Transformer
• Multi-feature Fusion