Tech Report
Integration of Information Extraction with an Ontology
This paper describes the integration of an ontology with an information extraction (IE) tool. Our main goal is extract knowledge from text to populate the ontology, and so alleviate the problem of ontology maintenance. The IE tool extracts information using partial parsing and machine learning techniques. Our domain of study is ``KMi Planet'', a Web-based news server that helps to communicate relevant information between members in our institute. Currently our system
finds instances of classes or subclasses. Although in the future we expect to create new classes and subclasses from new concepts appearing in text.