VOYAGE: A Large Collection of Vocabulary Usage in Open RDF Datasets

Shared vocabularies facilitate data integration and application interoperability on the Semantic Web. An investigation of how vocabularies are practically used in open RDF data, particularly with the increasing number of RDF datasets registered in open data portals, is expected to provide a measurement for the adoption of shared vocabularies and an indicator of the state of the Semantic Web. To support this investigation, we constructed and published VOYAGE, a large collection of vocabulary usage in open RDF datasets. We built it by collecting 68,312 RDF datasets from 517 pay-level domains via 577 open data portals, and we extracted 50,976 vocabularies used in the data. We analyzed the extracted usage data and revealed the distributions of frequency and diversity in vocabulary usage. We particularly characterized the patterns of term co-occurrence, and leveraged them to cluster vocabularies and RDF datasets as a potential application of VOYAGE. Our data is available from Zenodo at https://zenodo.org/record/7902675. Our code is available from GitHub at https://github.com/nju-websoft/VOYAGE.