A reproducible framework to assess the quality of linked open data in GLAM datasets

Over the last decade GLAM (Galleries, Libraries, Archives and Museums) organizations have been exploring new ways to make their content available using the Semantic Web and Linked Open Data (LOD). The growing number of GLAM institutions converting their collections to LOD, coupled with the increasing demand for high-quality data, has made the assessment of LOD quality a critical concern. In addition, there has been a significant increase in the global interest among researchers in reproducible research, a cornerstone of Open Science, requiring code to generate the experimental results. This study aims to present a reproducible framework to assess LOD quality within the GLAM sector. Based on the literature, a number of data quality criteria were established including 4 dimensions and 18 criteria. Subsequently, four LOD repositories were assessed according to these criteria. The assessments revealed that the LOD repositories performed well on accessibility, while they did not yield satisfactory results for other criteria, such as contextual information. These results can serve as a benchmark for other LOD repositories. Additionally, the study provides a detailed analysis that could be beneficial for other organizations and researchers interested in making their digital collections accessible and reusable as LOD. The study concludes by identifying further research based on the implementation in real practice for data spaces, the Cultural Heritage Cloud (ECCCH) and FAIR advancement in GLAMs.