A reliable linguistic filter for Farsi term extraction

Ontology learning has four main successive mining phases: term, concept, relation and axiom extraction. Since term extraction (TE) is the foundation of the ontology learning process, its inaccuracy will propagate in the following layers. Moreover, for any general-purpose ontology system some domain independent term extraction is needed. To achieve domain independent and precise TE in Farsi the present paper has presented a new hybrid system. It is based on the combination of a linguistic filter on Ezafe construction in Farsi and a statistical filter on C-value algorithm.