AKT EPrint Archive

Ontosophie: A Semi-Automatic System for Ontology Population from Text.

Celjuska, Mr David and Vargas-Vera, Dr Maria (2004) Ontosophie: A Semi-Automatic System for Ontology Population from Text.. In Proceedings International Conference on Natural Language Processing ICON 2004., Hyderabad, India..

This document is not available online.

This paper describes a system for semi-automatic population of ontologies with instances from unstructured text. It is based on supervised learning, learns extraction rules from annotated text and then applies those rules on new articles to populate the ontology. Hence, the system classifies stories and populates a hand-crafted ontology with new instances. It is based on three components: Marmot - a natural language processor; Crystal - a dictionary induction tool; and Badger - an information extraction tool. A part of the entire cycle is a user who accepts, rejects or modifies extracted and suggested instances to be populated. A description of experiments performed with a text corpus consisting of 91 articles is given. The results complete the paper and support the hypothesis that assigning a rule confidence value to each extraction rule improves the performance.

Keywords:Information Extraction, Ontology Population
Subjects:AKT Challenges > Knowledge acquisition
ID Code:416
Deposited By:Vargas-Vera, Dr Maria
Deposited On:08 June 2005
Alternative Locations:http://kmi.open.ac.uk/publications/pdf/kmi-04-19.pdf

Contact the site administrator at: hg@ecs.soton.ac.uk