From latent factors to language: a user study on LLM-generated explanations for an inherently interpretable matrix-based recommender system
This work addresses the need for user-centered explanations in explainable AI for recommender systems, though it is incremental as it applies existing LLM techniques to a known interpretable model.
The study investigated whether large language models (LLMs) could generate effective explanations for an interpretable matrix-based recommender system, finding through a user study with 326 participants that all explanation types were generally well-received with moderate differences between strategies.
We investigate whether large language models (LLMs) can generate effective, user-facing explanations from a mathematically interpretable recommendation model. The model is based on constrained matrix factorization, where user types are explicitly represented and predicted item scores share the same scale as observed ratings, making the model's internal representations and predicted scores directly interpretable. This structure is translated into natural language explanations using carefully designed LLM prompts. Many works in explainable AI rely on automatic evaluation metrics, which often fail to capture users' actual needs and perceptions. In contrast, we adopt a user-centered approach: we conduct a study with 326 participants who assessed the quality of the explanations across five key dimensions-transparency, effectiveness, persuasion, trust, and satisfaction-as well as the recommendations themselves. To evaluate how different explanation strategies are perceived, we generate multiple explanation types from the same underlying model, varying the input information provided to the LLM. Our analysis reveals that all explanation types are generally well received, with moderate statistical differences between strategies. User comments further underscore how participants react to each type of explanation, offering complementary insights beyond the quantitative results.