A Novel Approach for Ontology Maintenance Based on Web Data

Ontology is composed of a set of concepts and their mutual relations, thus it is the basis for the search engine to make semantically reasoning work. The accuracy of the concepts and other information in ontology is directly related to performance of a search engine. Thus the maintenance and optimization method for ontology is vital for the ontology construction and utilization, especially when the data source of ontology is now from millions of unstructured web pages data. There are many optimization approaches for ontology in the literature, but the automatic method is seldom found, and the most commonly used method is still artificial maintenance which is slow and inconvenient. Based on probability and lexical analysis, we proposed a hybrid ontology automatic maintenance approach in this paper. Experimental results show the effectiveness of this method.