LODAC Museum: Providing Linked Open Data for Academic Resources
As Japan’s only general academic research institution seeking to create future value in the new discipline of informatics, the National Institute of Informatics (NII) seeks to advance integrated research and development activities in information-related fields, including networking, software, and content. As an inter-university research institute, NII promotes the creation of a state-of-the-art academic-information infrastructure (the Cyber Science Infrastructure, or CSI) that is essential to research and education within the broader academic community, with a focus on partnerships and other joint efforts with universities and research institutions throughout Japan, as well as industries and civilian organizations. Ontotext AD is such a partner supporting NII goals with its semantic technology.
NII implemented the LOD.AC project with the goal to create and provide Linked Data for Japanese academic resources including museum information. Future plans included contribution of more generic Linked Data including Japanese contents such as dbpedia.jp.
LOD.AC Museum tackled several challenges including:
- Aggregate cultural heritage and museum information from all museums in Japan
- Present it as a website for researchers and other audiences
- Integrate natural language search and provide reliable SPARQL endpoint
Many Japanese museums have digitized their museum collections. This data is scraped by LOD.AC and mapped to RDF in CIDOC CRM. Associated artists and artworks from different museum collections are associated and integrated data views are provided. The result is LOD.AC-Museum, including web presentation, natural language search and SPARQL endpoint.
The project uses OWLIM semantic database to store the datasets and provide a backend to the public demo website http://lod.ac/. See the paper Producing and Consuming Linked Open Data on Art with a Local Community (COLD 2012) for more information.
Database configuration is BigOWLIM 3.5 on a 8 cores, 72GB server covering 19 sources with 700 000 entities and 15 million triples.
LODAC Applications include:
Map Demo – integration of LODAC, DBPedia
Yokohama Art Spot