Explainable Recommendation via Interpretable Feature Mapping and Evaluation of Explainability

TL;DR

This paper introduces a feature mapping method for recommender systems that balances accuracy and explainability without relying on metadata, using novel evaluation metrics for aspect-level explanations.

Contribution

A new feature mapping approach that enhances both recommendation accuracy and explainability by minimizing prediction and interpretation losses simultaneously.

Findings

Achieves high recommendation accuracy and explainability

Eliminates the need for metadata in explanations

Proposes new metrics for evaluating aspect-level explanations

Abstract

Latent factor collaborative filtering (CF) has been a widely used technique for recommender system by learning the semantic representations of users and items. Recently, explainable recommendation has attracted much attention from research community. However, trade-off exists between explainability and performance of the recommendation where metadata is often needed to alleviate the dilemma. We present a novel feature mapping approach that maps the uninterpretable general features onto the interpretable aspect features, achieving both satisfactory accuracy and explainability in the recommendations by simultaneous minimization of rating prediction loss and interpretation loss. To evaluate the explainability, we propose two new evaluation metrics specifically designed for aspect-level explanation using surrogate ground truth. Experimental results demonstrate a strong performance in both recommendation and explaining explanation, eliminating the need for metadata. Code is available from https://github.com/pd90506/AMCF.