Extracted with named-entity recognition
DocuMine offers a named-entity recognition class (Ner, NER) that allows you to handle a group of named-entity recognition entities to automatically identify named entities like persons, organizations, places etc.
Named entities in DocuMine:
Named entity | Technical name |
---|---|
Person | CBI_author |
Company or organization | ORG |
Street or part of an address | STREET |
Hous number of an address | CARDINAL |
Postal code | POSTAL |
Country | COUNTRY |
City | CITY |
Country | COUNTRY |
Federal State | STATE |
The following example rule corresponds to the best practice of combining an AI and a dictionary-based approach. It targets Ner entities of the type “country” within sections containing the string “PERFORMING LABORATORIES” and creates “Laboratory Country” entities for them.
rule "DOC.7.0: Performing Laboratory (Country)" when nerEntities: NerEntities(hasEntitiesOfType("COUNTRY")) $section: Section(containsString("PERFORMING LABORATORY:")) then nerEntities.streamEntitiesOfType("COUNTRY") .filter(nerEntity -> $section.getTextRange().contains(nerEntity.textRange())) .map(nerEntity -> entityCreationService.byNerEntity( nerEntity, "laboratory_country", EntityType.ENTITY, $section)) .forEach(entity -> entity.apply("DOC.7.1", "Performing Laboratory found")); end
Notice
For further information about the methods listed in the table, please refer to the Javadoc.