A new semantic model for domain-ontology learning

Abstract

Nowadays, ontology has been widely used in computer science. With the rapid development of it, Ontology Learning (OL) becomes one of the main research topics. However, with the current technology, all existing OL (from unstructured data source) methods rely on predefined knowledge about a specific domain. In this case, it’s more like an ontology enrichment process rather than a learning process. In this paper, a new learning model will be introduced. Rather than working on language level, this new approach focusing on structure level and trying to identify why and how a particular structure can represent a specific thing. Therefore, it allows the OL process not to rely on the predefined knowledge about a specific domain which people want to build an ontology for.