Cross-domain correspondences for explainable recommendations
View / Open Files
Authors
Stockdill, A
Raggi, D
Jamnik, M
Garcia, GG
Sutherland, HEA
Cheng, PCH
Sarkar, A
Publication Date
2020Journal Title
CEUR Workshop Proceedings
Conference Name
ExSS-ATEC Workshop at IUI 2020
Series
CEUR Workshop Proceedings
ISSN
1613-0073
Publisher
CEUR-WS.org
Volume
2582
Type
Conference Object
This Version
VoR
Metadata
Show full item recordCitation
Stockdill, A., Raggi, D., Jamnik, M., Garcia, G., Sutherland, H., Cheng, P., & Sarkar, A. (2020). Cross-domain correspondences for explainable recommendations. CEUR Workshop Proceedings, 2582 https://doi.org/10.17863/CAM.81027
Abstract
Humans use analogies to link seemingly unrelated domains. A mathematician might discover an analogy that allows them to use mathematical tools developed in one domain to prove a theorem in another. Someone could recommend a book to a friend, based on understanding their hobbies, and drawing an analogy between them. Recommender systems typically rely on learning statistical correlations to uncover these cross-domain correspondences, but it is difficult to generate human-readable explanations for the correspondences discovered. We formalise the notion of 'correspondence' between domains, illustrating this through the example of a simple mathematics problem. We explain how we might discover such correspondences, and how a correspondence-based recommender system could provide more explainable recommendations.
Identifiers
External DOI: https://doi.org/10.17863/CAM.81027
This record's URL: https://www.repository.cam.ac.uk/handle/1810/333611
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.