Robust ontology acquisition from machine-readable dictionaries
Date of Issue2005
International Joint Conference on Artificial Intelligence (2005)
School of Humanities and Social Sciences
In this paper, we outline the development of a system that automatically constructs ontologies by extracting knowledge from dictionary definition sentences using Robust Minimal Recursion Semantics (RMRS), a semantic formalism that permits underspecification. We show that by combining deep and shallow parsing resources through the common formalism of RMRS, we can extract ontological relations in greater quality and quantity. Our approach also has the advantages of requiring a very small amount of rules and being easily adaptable to any language with RMRS resources.
© 2005 Proceedings of the International Joint Conference on Artificial Intelligence.