Towards a Comprehensive Assessment of the Quality and Richness of the Europeana Metadata of food-related Images

Yalemisew Abgaz, Amelie Dorn, Jose Luis Preza Diaz, Gerda Koch


Abstract
Semantic enrichment of historical images to build interactive AI systems for the Digital Humanities domain has recently gained significant attention. However, before implementing any semantic enrichment tool for building AI systems, it is also crucial to analyse the quality and richness of the existing datasets and understand the areas where semantic enrichment is most required. Here, we propose an approach to conducting a preliminary analysis of selected historical images from the Europeana platform using existing linked data quality assessment tools. The analysis targets food images by collecting metadata provided from curators such as Galleries, Libraries, Archives and Museums (GLAMs) and cultural aggregators such as Europeana. We identified metrics to evaluate the quality of the metadata associated with food-related images which are harvested from the Europeana platform. In this paper, we present the food-image dataset, the associated metadata and our proposed method for the assessment. The results of our assessment will be used to guide the current effort to semantically enrich the images and build high-quality metadata using Computer Vision.
Anthology ID:
2020.ai4hi-1.5
Volume:
Proceedings of the 1st International Workshop on Artificial Intelligence for Historical Image Enrichment and Access
Month:
May
Year:
2020
Address:
Marseille, France
Venues:
AI4HI | LREC | WS
SIG:
Publisher:
European Language Resources Association (ELRA)
Note:
Pages:
29–33
Language:
English
URL:
https://www.aclweb.org/anthology/2020.ai4hi-1.5
DOI:
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/2020.ai4hi-1.5.pdf