An Iterative Method of Extracting Chinese ISA Relations for Ontology Learning

Abstract

Automatic acquisition of ISA relations is a basic problem in knowledge acquisition from text. We present an iterative method extracting ISA relations from large Chinese free text for ontology learning. Firstly, it initially discovers a set of sentences using several special Chinese lexico-syntactic patterns from free text corpus. Secondly we combine outside layer removal and inside layer gathering for acquiring concepts of constituting ISA relation. Finally, ISA relations are verified with multiple features. Extracted ISA relations will be selected for new relation extracting cycle. Experimental results demonstrate good performance of the method for extracting ISA relation from large Chinese corpus.