Skip to content

ISIDORE

Tobias Steiner edited this page Apr 30, 2021 · 4 revisions

Run by Huma-Num, ISIDORE is a search engine providing access to digital data from the Humanities and Social Sciences (SSH), including >1,3 mio OA books and book chapters.

ISIDORE harvests records, metadata and full text from electronic publications, corpora, databases and scientific news, accessible on the web and offered in open and interoperable standards. ISIDORE harvests data in French (produced in France or in the French-speaking world), English and Spanish. Once harvested, this information is enriched in three languages (English, Spanish and French) by cross-referencing with professional repositories (vocabulary lists, thesauri, repositories). These repositories are produced either by the scientific community (GeoEthno, Pactols, etc.) or by major institutions in the field of higher education and research (Rameau, LCSH, BNE, Gemet, Lexvo, GeoNames, etc.). Multilingual enrichments allow data to be linked together. This information is entry points to the full text, which is also indexed when possible.

APIs

ISIDORE APIs are available at https://api.isidore.science through the GET method on HTTP or HTTPS. They provide a fast, accurate and reliable service for querying ISIDORE data with extensive search functionalities (autocompletion, spelling correction, multi-criteria, boolean and faceted searches, sorting, aggregation of responses, etc.).

Each request to the engine is submitted through a URI pointing to a specific web service. The response is a stream in XML (default format) or JSON format.

A SPARQL endpoint is available at https://api.isidore.science/sparql with documentation at https://isidore.science/sqe

Clone this wiki locally