Responsible AI in Personalised Movie Recommender Systems for the Hearing Impaired Community

Vachhani, Roshni; Hadzidedic, Suncica

Abstract

The aim of this paper is to develop a personalised recommender system (RS) in the movie domain - MRSystem - with a focus on Responsible AI for the hearing impaired community. There is currently no movie RS that looks at protected characteristics when generating recommendations to ensure that no user is disadvantaged. MRSystem was developed to look at data from MovieLens and Netflix databases, identify user preferences and any previous patterns on Closed Captions, and then provide personalised recommendations that match patterns identified using conventional RS techniques and Responsible AI principles. The results from the user study and offline evaluations we performed highlight the need for disability-aware RSs that account for Responsible AI principles.

Citation

Vachhani, R., & Hadzidedic, S. (2024, September). Responsible AI in Personalised Movie Recommender Systems for the Hearing Impaired Community. Presented at 2024 Intelligent Systems Conference (IntelliSys), Amsterdam, The Netherlands

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2024 Intelligent Systems Conference (IntelliSys)
• Start Date: Sep 5, 2024
• End Date: Sep 6, 2024
• Acceptance Date: Jul 10, 2024
• Online Publication Date: Aug 1, 2024
• Publication Date: Aug 1, 2024
• Deposit Date: Oct 9, 2024
• Publicly Available Date: Oct 9, 2024
• Print ISSN: 2367-3370
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Pages: 195-214
• Series Title: Lecture Notes in Networks and Systems
• Series ISSN: 2367-3389
• Book Title: Intelligent Systems and Applications Proceedings of the 2024 Intelligent Systems Conference (IntelliSys) Volume 4
• ISBN: 9783031663352
• DOI: https://doi.org/10.1007/978-3-031-66336-9_15
• Public URL: https://durham-repository.worktribe.com/output/2951183