Automation of ontology development from domain specific concept taxonomy

Currently, much of the information on the Web is described using only natural language, which can be seen as a major obstacle in developing the Semantic Web. Since the annotations describing different resources are one of the key components of the Semantic Web, we need to create them easy to use and cost-effective ways. However, various systems for creating annotations have been developing; there seem to be a lack of systems that can be easily used by annotators unfamiliar with the technical side of the Semantic Web. In this paper, extraction possible concepts relating to a particular domain are stored in concept taxonomy and describe how agents can contribute to develop ontology automatically. We have developed an annotation system supporting distributed collaboration with concept ontology, and hiding the complexity of the annotation scheme. Given concept taxonomy for a particular application domain, software agent can represent and organize data pertinent to this domain more effectively than without any prior knowledge of the relationships between concepts. This system adapts flexibly to different metadata schemas, which makes it suitable for different applications. Support for using ontologies is based on agents and concept searching.