| Ontology is a formal expression of a set of concepts and their mutual relationships in certain areas. As a kind of knowledge expression method, ontology has been widely used in many fields, such as Tim Berners-Lee’s Semantic Web conception and biomedical informatics. Biomedicine is one of the most active and frontier fields of ontology application, and there have been several successful application of ontology. One of the representatives is the UMLS(Unified Medical Language System), which is designed and updated by the U.S. National Library’of Medicine. As a huge ontology. UMLS contains millions of biomedical concepts and relationships, while the concepts are too vast and the relationships are not clear enough to support further knowledge acquisition and information processing. Meanwhile, the fact that traditional ontology construction relies on manual operation with low automation and customization to a certain extent limits the application of ontology.Based on these situations, an idea of specific sub-domain ontology automatic construction is proposed in this thesis. A small and detailed ontology of sub-domain that users focus on is to be constructed, so that it facilitates knowledge acquisition and further applications. In this paper, the author designs and realizes a biomedical specific domain ontology automatic construction system. The system takes the concepts in UMLS as the universal concept set of the ontology, extracts concepts from domain text provided by the customer, constructs an ontology descripted in OWL, and provides a graphical editor. As an application of this system, the author also realizes a function of searching keywords recommending based on semanteme.The significance of this paper is that it completely realizes a biomedical ontology automatic construction system with friendly interface, and makes an attempt in ontology application. The result of experiment shows that the system can construct biomedical ontologies on customer requirements well with efficiency. |
PDF Full Text Request