paper45.pdf (116.97 kB)
LOD Surfer Metadata: Essential LOD catalogue data for traversing life-science LOD amongst multiple SPARQL endpoints
posterposted on 2018-12-03, 13:36 authored by Norio Kobayashi, Yasunori YamamotoYasunori Yamamoto
Researchers in the life sciences are advanced developers and publishers of Linked Open Data (LOD). Because the datasets generated by various apparatuses and experimental methods are heterogeneous and diverse, LOD descriptions involve not only widely and commonly used but also uses’ original data classes and ontology terms. To resolve this problem, we have revised the currently proposed LOD Surfer Metadata, which describe detailed LOD graph structures such as class--class relationships and statistics, by introducing upper-level conceptual classes based on Medical Subject Headings (MeSH) terms. With these descriptions, users can easily discover and integrate multiple LOD over classes of different ontologies. Furthermore, the availability and frequency of data updates can be checked by a public SPARQL endpoint monitoring service. Accordingly, we have been developing a practical LOD-discovery service for integrative data analysis such as SPARQL federate searching. This poster presents our developmental progress on the LOD Surfer Metadata and LOD discovery service.
JSPS KAKENHI grant numbers 17K00434 and 17K00424