Every organisation has unstructured resources which often contain indispensable data. In order to obtain value from this data it must be converted from plain meaningless text to enriched resources with a meaning in a context. We create Linked Data from your E-mail, Twitter, websites and documents, in English as well as in Dutch.
... Our experience with natural language analysis has proven indispensable in many projects. We apply natural language analysis techniques to process all steps necessary to create knowledge-rich RDF from plain data. Concept recognition is a technique to find meaningful concepts in unstructured content, which are used to structure and enrich content with metadata and to discover meaningful links with other resources. As a result, unstructured data such as E-mail, Twitter, websites, or documents, is provided with meaningful metadata and links, and is added to your Linked Data set.
Concept recognition requires a knowledge model. Where possible we use standard vocabularies and where necessary we create an ontology. We create ontologies semi-automatically from structured and unstructured resources and where necessary we create ontologies from tacit knowledge that remains in the heads of experts using knowledge elicitation techniques.