Ontology Construction Method Of Field-based Knowledge Map

With depth study of computer technology and the rapid development of Internet, more and more information appears in front of people, which makes it difficult for the user to quickly find relevant informations. Ontology as a description of the objective reality of the world, providing a common understanding of knowledge, has been widely applied to computer science and information. The way to build ontology has become a new hotspot, in the different specific areas ontology modeling have a different methods. Specific projects in foreign countries made a number of possible ways, and the present ontology building about Chinese is still in the exploratory stages. As knowledge support of information extraction, semantic web and a series of natural language processing applications, it is the general concept of hierarchical model in a particular area or scene of refinement and generalization, directly determine the precision and recall of the various applications based on it. Construction of the existing manual method is time consuming and labor-intensive, automated extraction of knowledge that is not clear enough, this makes the final ontology semantic description ambiguity, led to the results based on it not effect enough.This article describes starting with semantic description, proposed a domain ontology construction method based on knowledge graphs. A domain ontology was constructed based on the mature idea of software engineering by using the "HowNet" as the semantic knowledge resources, and knowledge graphs as a semantic representation. It has a clear structure and semantic meaning. It can provide effective supports for some applications such as Semantic Web and information extraction. This paper introduced the concept of ontology, the standard of design, the process of modeling, and future prospects of ontology portability.In the end of this paper, we compared with the traditional modeling methods and given the evaluation results. Since this article emphasizes the representation of knowledge, making domain ontology provides a good understanding of knowledge, especially for knowledge with ambiguity through word segmentation, disambiguation, tagging, the so messy unstructured text to become the order ones. This method was proved to be more effective in uncertain knowledge processing by experiment simulations, improved the precision and recall.