Ontosophie: A Semi-Automatic System for Ontology Population from Text David Celjuska MSc
This event took place on 6th December 2004 at 12:30pm (12:30 GMT)
Knowledge Media Institute, Berrill Building, The Open University, Milton Keynes, United Kingdom, MK7 6AA
In this talk I will describe Ontosophie, a system for semi-automatic population of ontologies with instances from unstructured text. Extraction rules are generated from annotated text using supervise learning techniques. These rules are then applied to 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.
In the talk I will address the major challenges and introduce confidence values that we implemented in the system to enhance its performance. Different methods of confidence computation will be given and their results compared on a text corpus consisting of KMi news articles.
Finally, the presentation will be followed with a brief demonstration of Ontosophie.
The talk is being hosted by Dr. Maria Vargas-Vera from KMi.
Click below to play the event (63 minutes)