Temporal Meta-path Guided Explainable Recommendation

TL;DR

This paper introduces a novel explainable recommendation method using item-item path modeling and attention mechanisms on dynamic knowledge graphs, outperforming existing RNN-based approaches.

Contribution

It proposes a simple neural network framework leveraging path-based context for dynamic recommendations, offering improved interpretability and performance.

Findings

Achieves state-of-the-art results on three benchmark datasets

Outperforms recent strong baselines in recommendation accuracy

Provides explainability through path-based modeling

Abstract

This paper utilizes well-designed item-item path modelling between consecutive items with attention mechanisms to sequentially model dynamic user-item evolutions on dynamic knowledge graph for explainable recommendations. Compared with existing works that use heavy recurrent neural networks to model temporal information, we propose simple but effective neural networks to capture user historical item features and path-based context to characterise next purchased item. Extensive evaluations of TMER on three real-world benchmark datasets show state-of-the-art performance compared against recent strong baselines.