Research on domain ontology semi-automatic construction based on FCA and Jena

To aim at solving the difficult problems of discovering the implicit knowledge and coding for ontology inefficiently in ontology construction, this paper proposes a semi-automatic approach to construct the domain ontology based on FCA and Jena. At first, formal context is built up by clustering the instance sets and the attribute sets of the domain concept repeatedly and then the concept lattice is built up. Secondly, the domain ontology needs to be visualized. Furthermore, the potential concepts and relations among concepts are found to form the ontology trunk. At last, the ontology trunk is enriched by adding the corresponding attribute values. After that, the OWL is selected to code and formalize the above original ontology based on Jena in order to complete the construction of domain ontology. Taking the Yunnan tourism domain as background, the tourism ontology prototype system is constructed, which demonstrates the validity of the domain ontology semi-automatic construction approach proposed in this paper, and some evaluations on field portability, ontology building efficiency and degree of automation have good results. Compared with artificial construct results, it shows the construction validity.