Sustainable transparency on recommender systems: Bayesian ranking of images for explainability

• Published in: Information fusion 2024-11, Vol.111, p.102497, Article 102497
• Main Authors: Paz-Ruza, Jorge, Alonso-Betanzos, Amparo, Guijarro-Berdiñas, Bertha, Cancela, Brais, Eiras-Franco, Carlos
• Format: Article
• Language: English
• Subjects: Dyadic data; Explainable Artificial Intelligence; Explainable recommendations; Frugal AI; Machine Learning; Recommender systems
• ISSN: 1566-2535; 1872-6305
• DOI: 10.1016/j.inffus.2024.102497

Recommender Systems have become crucial in the modern world, commonly guiding users towards relevant content or products, and having a large influence over the decisions of users and citizens. However, ensuring transparency and user trust in these systems remains a challenge; personalized explanations have emerged as a solution, offering justifications for recommendations. Among the existing approaches for generating personalized explanations, using existing visual content created by users is a promising option to maximize transparency and user trust. State-of-the-art models that follow this approach, despite leveraging highly optimized architectures, employ surrogate learning tasks that do not efficiently model the objective of ranking images as explanations for a given recommendation; this leads to a suboptimal training process with high computational costs that may not be reduced without affecting model performance. This work presents BRIE, a novel model where we leverage Bayesian Pairwise Ranking to enhance the training process, allowing us to consistently outperform state-of-the-art models in six real-world datasets while reducing its model size by up to 64 times and its CO2 emissions by up to 75% in training and inference. •A new model named BRIE is proposed for visual-based recommendation explainability.•Existing models use highly optimized architectures but suboptimal training policies.•BRIE uses a more adequate learning objective and use of training data.•BRIE outperforms performance of state-of-the-art models in benchmark datasets.•BRIE can reduce model size by up to 64 times and carbon emissions by up to 75%.