Impact of Tone-Aware Explanations in Recommender Systems
Authors: A Okoso, K Otaki, S Koide, Y Baba
Published: 2025
Publication: ACM Transactions on Recommender Systems, 2025•dl.acm.org
The study demonstrates that tailoring the tone of textual explanations in recommender systems to domains and user attributes, such as age and personality traits, can enhance users' perceptions and engagement.
Methods: Two online user studies: (1) 470 participants evaluated synthetic explanations with six tones across three domains (movies, hotels, and home products), (2) 103 participants engaged with a real-world dataset from the hotel domain using a personalized recommender system.
Key Findings: The perceived effects of different textual explanation tones on users, examined across domains (movies, hotels, home products) and user attributes (e.g., age, personality traits).
Limitations: Results may not generalize fully beyond the tested domains as only three were explored; real-world datasets were only utilized in the hotel domain, limiting applicability to other domains.
Institution: Toyota Central R and D Labs, Toyota
Research Area: Human-Computer Interaction (HCI)
Discipline: Machine Learning, Artificial Intelligence
Sample Size: 573 participants
Citations: 13
DOI: https://dl.acm.org/doi/10.1145/3718101