Visualization for Recommendation Explainability: A Survey and New Perspectives
Mohamed Amine Chatti, Mouadh Guesmi, Arham Muslim
TL;DR
This survey addresses the need for visual explanations in recommender systems by adopting Munzner's What-Why-How visualization framework to classify explanatory visualizations along four dimensions: aim, scope, method, and format. Analyzing 33 tools from 2000–2021, the authors find that most visual explanations target justification of recommendations and output-level explanations, while algorithmic transparency and input-model explanations are comparatively underexplored. They report a dominance of content-based and social/hybrid methods, with node-link diagrams and bar charts as the most common visual idioms, and emphasize the importance of interactive, layered explanations to support varied user needs. The paper offers design guidelines and identifies research gaps, including the need for theory-informed visualization designs, systematic evaluation of different aims, and better integration of visualization guidelines from the information visualization field to improve explainable recommender systems. Overall, the work provides a structured pathway for researchers and practitioners to design, evaluate, and implement effective visually explainable recommendations that balance simplicity, detail, and user control.
Abstract
Providing system-generated explanations for recommendations represents an important step towards transparent and trustworthy recommender systems. Explainable recommender systems provide a human-understandable rationale for their outputs. Over the last two decades, explainable recommendation has attracted much attention in the recommender systems research community. This paper aims to provide a comprehensive review of research efforts on visual explanation in recommender systems. More concretely, we systematically review the literature on explanations in recommender systems based on four dimensions, namely explanation goal, explanation scope, explanation style, and explanation format. Recognizing the importance of visualization, we approach the recommender system literature from the angle of explanatory visualizations, that is using visualizations as a display style of explanation. As a result, we derive a set of guidelines that might be constructive for designing explanatory visualizations in recommender systems and identify perspectives for future work in this field. The aim of this review is to help recommendation researchers and practitioners better understand the potential of visually explainable recommendation research and to support them in the systematic design of visual explanations in current and future recommender systems.