International Journal of Computer Science Issues

More than a traditional journal...

In recent years the development of Ontologys, leads to explicit formal specifications of the terms in the domain and relations among them. The construction of Ontology often requires a domain specific corpus in conceptualizing the domain knowledge. It is an indispensible task to identify a list of significant terms for constructing a Structured Ontology. In this paper, we investigate the use of Semantic Similarity-based metrics for term recognition and extraction, for ontology construction from the text document. The methodology uses Taxonomy and Wikipedia to reinforce the automatic term recognition and extraction from structured documents. It is done with the assumption of candidate terms for a topic are often associated with its topic-specific keywords through Semantic Similarity-based metrics by making use of WordNet. A hierarchical relationship of super-topics and sub-topics is defined by Taxonomy, meanwhile, Wikipedia is used to provide a semantic relationship and background knowledge for topics that are defined in the Taxonomy to surpervise the term recognition and extraction. Experimental results show that the proposed methodology is viable to be applied in a small corpus supporting Ontology construction, which renders the foundation for higher recall and precision, when compared with existing methodologies.