An Introduction to Hasti: An Ontology Learning System

M. Shamsfard and A.A. Barforoush (Iran)


Ontology, Knowledge Acquisition, Natural LanguageProcessing, Lexicon, Machine Learning.


In recent years ontologies have been widely used in information systems. The major problems in building and using ontologies are the bottleneck of knowledge acquisition and time-consuming construction and integration of various ontologies. Meanwhile moving toward automation of ontology building is a solution. This paper introduces Hasti, an automatic ontology building system. Hasti extracts lexical and ontological knowledge from Persian (Farsi) written texts. Its lexicon is near empty initially and will grow gradually by learning new words. Its ontology is a small kernel at the beginning. The initial kernel has the essential Meta knowledge (primitive concepts and operators) to build an ontology. It is language neutral and domain independent. The built ontology will be dynamic, flexible to changes and automatic expandable. It is formed of concepts, taxonomic and non-taxonomic conceptual relations between concepts, and axioms, all as ontology elements. The ontology building process works in a middle-out manner. In this paper we will describe Hasti, its features and its components.

Important Links:

Go Back