Making the Model Visible: Modelling Performance and Methodological Provenance in an AI-Assisted Digital Humanities Workflow
Journal of the Digital Humanities Association of Southern Africa (DHASA) · 20.11.2026 · 7 (2) · 96-105
DOI: 10.55492/dhasa.v7i02.7695 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the Digital Humanities Association of Southern Africa (DHASA), online erschienen am 20.11.2026. Verfasst von Sanjin Muftić. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
The South African College of Music (SACM) at the University of Cape Town holds a substantial collection of concert recordings and accompanying programme data. Preparing this material for publication through Ibali: Digital Collections UCT required relationships implicit in the original spreadsheets to be represented explicitly within Omeka S. This paper examines the development of a linked-resource model structured around four resource types: Concert, Performance, Work and Performance Credit. Particular attention is given to distinguishing a musical Work from an individual Performance, and to representing performer participation as a separate resource when information such as instrument, voice type or character belongs to the relationship itself. The paper also considers practical consequences of the model, including the use of stable identifiers across successive imports and the challenge of creating meaningful labels for relational resources within the user interface. The model was developed iteratively with generative AI used to explore structures, generate data-transformation methods and test possible representations. Drawing on this process, the paper argues that data modelling makes interpretive distinctions computationally visible, while significant human-AI modelling decisions should themselves form part of the methodological provenance through which digital humanities resources can later be understood and reused.
Zitieren
APA
Muftić, S. (2026). Making the Model Visible: Modelling Performance and Methodological Provenance in an AI-Assisted Digital Humanities Workflow. Journal of the Digital Humanities Association of Southern Africa (DHASA), 7(2), 96-105. https://doi.org/10.55492/dhasa.v7i02.7695
Vancouver
Muftić S. Making the Model Visible: Modelling Performance and Methodological Provenance in an AI-Assisted Digital Humanities Workflow. Journal of the Digital Humanities Association of Southern Africa (DHASA). 2026;7(2):96-105. doi:10.55492/dhasa.v7i02.7695
BibTeX
@article{Mufti2026,
author = {Muftić, Sanjin},
title = {Making the Model Visible: Modelling Performance and Methodological Provenance in an AI-Assisted Digital Humanities Workflow},
journal = {Journal of the Digital Humanities Association of Southern Africa (DHASA)},
year = {2026},
volume = {7},
number = {2},
pages = {96--105},
doi = {10.55492/dhasa.v7i02.7695}
}
EUVIMED