By combining our core ontology management, pattern recognition and language processing technologies we can provide mechanisms to extract useful entities such as company names, product names etc from documents.
The ontology provides the schema and rules for the extraction process. Documents are pre-processed to find potential matches and then the pattern recognition takes over to find good matches against the ontology. As with the categorisation technology the concepts (in this case matching entities) are returned with information indicating the relevance to the document.
The entity information is typically used in one of two ways. Firstly we may want to place the documents in an information retrieval system. Secondly we may want to use the entities to decide how information is routed or distributed to users in an alerting system. For example users may subscribe to documents about a particular company.