Do calibrated recommendations affect explanations? A study on post-hoc adjustments

Simple item page
dc.contributor.authorAtauchi, Paul Dany Floresen
dc.contributor.authorZanon, André Levien
dc.contributor.authorda Rocha, Leonardo Chaves Dutraen
dc.contributor.authorManzato, Marcelo Garciaen
dc.contributor.funderCoordenação de Aperfeiçoamento de Pessoal de Nível Superioren
dc.contributor.funderConselho Nacional de Desenvolvimento CientĂ­fico e TecnolĂłgicoen
dc.contributor.funderFundação de Amparo à Pesquisa do Estado de São Pauloen
dc.contributor.funderAmazon Web Servicesen
dc.date.accessioned2025-07-17T14:43:58Z
dc.date.available2025-07-17T14:43:58Z
dc.date.issued2025en
dc.description.abstractRecommender systems generate suggestions by identifying relationships among past interactions, user similarities, and item metadata. Recently, there has been an increased focus on evaluating recommendations based not only on accuracy but also on aspects like transparency and calibration. Transparency is important, as explanations can enhance user trust and persuasion, while calibration aligns users’ interests with recommendation lists, improving fairness and reducing popularity bias. Traditionally, calibration and explanation are applied in post-processing. Our study investigates two key research gaps: (1) the impact of graph embeddings in model-agnostic knowledge graph explanations, exploring their under-researched potential compared to syntactic approaches to produce meaningful explanations; and (2) the effect of calibration on recommendation explanations, assessing whether calibrated recommendation reordering influences the outcomes of explanation algorithms. We evaluate the quality of explanations using a set of metrics, such as diversity, which measures how well different interests of the user are covered; popularity, which assesses how well explanations avoid favoring already popular items; and recency, which examines the inclusion of recently interacted items. Our findings demonstrate that graph embedding methods are effective in generating high-quality explanations using these offline explanation metrics, and that post-hoc knowledge graph explanation algorithms are robust to calibration changes.en
dc.description.statusPeer revieweden
dc.description.versionPublished Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationAtauchi, P. D. F., Zanon, A. L., da Rocha, L. C. D. and Manzato, M. G. (2025) 'Do calibrated recommendations affect explanations? A study on post-hoc adjustments', Journal on Interactive Systems, 16(1), pp.441-460. https://doi.org/10.5753/jis.2025.5563en
dc.identifier.doi10.5753/jis.2025.5563en
dc.identifier.endpage460en
dc.identifier.issn27637719en
dc.identifier.issued1
dc.identifier.journaltitleJournal on Interactive Systemsen
dc.identifier.startpage441en
dc.identifier.urihttps://hdl.handle.net/10468/17725
dc.identifier.volume16
dc.language.isoenen
dc.publisherBrazilian Computing Societyen
dc.rights© 2025, Brazilian Computing Society. This work is licensed under a Creative Commons Attribution 4.0 International License.en
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCalibrationen
dc.subjectExplanationen
dc.subjectGraph Embeddingsen
dc.subjectRecommender Systemsen
dc.titleDo calibrated recommendations affect explanations? A study on post-hoc adjustmentsen
dc.typeArticle (peer reviewed)en
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
5563-Article Text-29732-1-10-20250629.pdf
Size:
2.49 MB
Format:
Adobe Portable Document Format
Description:
Published Version
Download
Collections
Insight Centre for Data Analytics - Journal Articles