Graphing else matters: exploiting aspect opinions and ratings in explainable graph-based recommendations

TL;DR

This paper introduces a novel graph embedding approach that combines ratings and aspect-based opinions from reviews to improve recommendation accuracy and provide explainable suggestions.

Contribution

It proposes a new method to exploit combined rating and opinion embeddings from graphs, enhancing recommendation quality and interpretability.

Findings

Outperforms baseline recommenders on Amazon and Yelp datasets.

Provides explanations based on aspect opinions in recommendations.

Enables visualization of user preferences and aspect importance.

Abstract

The success of neural network embeddings has entailed a renewed interest in using knowledge graphs for a wide variety of machine learning and information retrieval tasks. In particular, current recommendation methods based on graph embeddings have shown state-of-the-art performance. These methods commonly encode latent rating patterns and content features. Different from previous work, in this paper, we propose to exploit embeddings extracted from graphs that combine information from ratings and aspect-based opinions expressed in textual reviews. We then adapt and evaluate state-of-the-art graph embedding techniques over graphs generated from Amazon and Yelp reviews on six domains, outperforming baseline recommenders. Our approach has the advantage of providing explanations which leverage aspect-based opinions given by users about recommended items. Furthermore, we also provide examples of the applicability of recommendations utilizing aspect opinions as explanations in a visualization dashboard, which allows obtaining information about the most and least liked aspects of similar users obtained from the embeddings of an input graph.