Kepa Rodriguez, "Multilingual and Semantic Search and Indexing in the EHRI Project"
Digital Text Analysis Working Group
Meeting of the GCDH Digital Text Analysis Working Group, Heyne Haus Seminarraum 1
Metadata records stored in the metadata registry of the EHRI project (http://www.ehri-project.eu <http://www.ehri-project.eu/> ) are highly multilingual. Not only because metadata records are written in different languages, but because at the same time more than a language can be present inside of the same metadata record without any language coding mark-up.
In this talk I will present how the EHRI project addresses the problem of multilinguality of the data in the information retrieval process with emphasis in the use of domain knowledge, control vocabularies and ontologies in the indexing process. Finally I will give some hints about a semantic enrichment strategy in the query time, which is helpful to extend the results of the search.
Presentation Slides: PDF