Lacking in context or widening access: Experiments with AI in cataloguing the Collecting Social Photo images
What are the potential benefits of using AI technology to support the cataloguing of our cultural collections? Can the technology be trusted to accurately record our cultural heritage, or is human intervention still needed. This paper unpacks the findings of an experiment using three Image recognition tools to catalogue 50 photographs from the Collecting Social Photo project. The paper argues that AI tools still have some way to go before they can understand context, but could be useful as tools to add additional tags to already catalogued collections to increase online discoverability and widen access.
Arran Rees
Since 2017 Arran has been collaborating closely with the Collecting Social Photo project, with the outcomes of the collaboration feeding both into the project and his PhD research. Arran is Doctoral Researcher at the School of Fine Art, History of Art and Cultural Studies at the University of Leeds. His research is interested in the impact collecting from social media will have on establsihed museum collections management standards. Arran started his career working in museum after completing his MA in Museum Studies at the University of Leicester. Working in both curatorial and collections management roles for seven years in museums across South Wales, and later at the Victoria and Albert Museum.