AOTree: Aspect Order Tree-based Model for Explainable Recommendation

TL;DR

This work addresses the need for explainable recommendations by incorporating Order Effects Theory, which posits that decision factors are considered in a specific sequence. It introduces AOTree, a three-stage model combining a personalized Aspect Order Tree (AOTree) with an Aspect Order Generator and a prediction module that uses position-aware self-attention to forecast ratings. Empirical results across five public datasets show that AOTree improves both predictive accuracy (lower MSE, higher NDCG) and explainability (higher aspect-order coverage and alignment with ground-truth sequences) compared with strong baselines. The method offers a practical path toward human-simulable explanations in recommendations, while acknowledging data sparsity and potential ethical considerations around filter bubbles.

Abstract

Recent recommender systems aim to provide not only accurate recommendations but also explanations that help users understand them better. However, most existing explainable recommendations only consider the importance of content in reviews, such as words or aspects, and ignore the ordering relationship among them. This oversight neglects crucial ordering dimensions in the human decision-making process, leading to suboptimal performance. Therefore, in this paper, we propose Aspect Order Tree-based (AOTree) explainable recommendation method, inspired by the Order Effects Theory from cognitive and decision psychology, in order to capture the dependency relationships among decisive factors. We first validate the theory in the recommendation scenario by analyzing the reviews of the users. Then, according to the theory, the proposed AOTree expands the construction of the decision tree to capture aspect orders in users' decision-making processes, and use attention mechanisms to make predictions based on the aspect orders. Extensive experiments demonstrate our method's effectiveness on rating predictions, and our approach aligns more consistently with the user' s decision-making process by displaying explanations in a particular order, thereby enhancing interpretability.

Figures (5)

• Figure 1: The $Consistency\_dis_{user/item}$ value among users/items. The horizontal axis represents the proportion of users/items under specific consistency difference.
• Figure 2: Example of the existence and variances of order effects in reviews.
• Figure 3: The algorithmic framework of AOTree.
• Figure 4: Example of calculating Split Value (SV).
• Figure 5: Hyperparameter analysis w.r.t. the max depth and latent dimension size on three datasets.