Recommender Systems in Museums and Galleries
With a fellow PhD student at the University of Manchester, we built a
recommender system to create personalised curations of museum content. We worked
working in collaboration with the Manchester and Whitworth Art Galleries and
the Smithsonian in the U.S.
The project involved training autoencoders to create feature representations
from the digital images in the collection and applying NLP techniques to
extract features from the artwork metadata. Further, it consisted of building
the recommendation system - a web app - to create a personalised exhibition
of the museum for each user, and carrying out a study to test its effectiveness.
Papers
- Recommending Art Online: Investigating Engagement and Interactions with a National Public Collection. L. Hughes-Nöhrer, J. Carlton, & C. Jay. Paper (pre-print)
- Machine Learning and Museum Collections: A Data Curation Conundrum. L. Nöhrer, J. Carlton, & C. Jay. International Conference on Emerging Technologies and the Digital Transformation of Museums and Heritage Sites (RISE IMET), Springer, 2021