Abstract
We describe OAT, a new information extraction system under development. It extracts (subject, predicate, object) triplets from natural language texts. It uses ontologies extensively: the results are saved in an ontology, and ontologies are used in the information extraction process itself. It is adaptable both to a domain of discourse and within a domain of discourse (finding new concepts). This paper concentrates on the requirements and architecture of OAT.
Original language | English |
---|---|
Title of host publication | Poster proceedings |
Subtitle of host publication | Industrial Conference on Data Mining, ICDM 2006. Leipzig, DE, 13 - 14 July 2006. |
Editors | Petra Perner |
Place of Publication | Leipzig |
Pages | 225-229 |
Publication status | Published - 2006 |
MoE publication type | Not Eligible |
Keywords
- information extraction
- ontology
- software requirements
- software architecture