In our continued efforts to build the Mikrokosmos machine translation
system, we have developed an elaborate semantic classification of
concepts, or ontology, to facilitate natural language
interpretation. The
K ontology is a large-scale
classification of a wide variety of meaning elements (called concepts)
with a rich interconnection of conceptual relations. We have shown
that the semantic classification we developed is of great value in
solving ambiguities and other related problems in natural language
processing. In addition, we have also developed a methodology for
developing the classification as well as a variety of state of the art
software tools for editing, browsing, and accessing large
classification systems. We are continuing to expand the coverage of
the ontology by acquiring more concepts. In addition, we are also
looking at new applications for the
K ontology.
The authors would like to thank Lori Wilson for doing the bulk of the work in ontology acquisition and other members of the Mikrokosmos team, including Stephen Beale, Evelyne Viegas, Victor Raskin, and Boyan Onyshkevych, for their valuable contributions to the work reported in this article.
Kavi Mahesh