A Research On Semantic Relevancy Computational Method For Text Based On Hypertension Domain Ontology

With the development of Internet technology,the amount of text information have exploded in the last several years.As an important carrier of information,text processing has been an important research subject in the field of information processing.The study of the semantic relevance calculation between texts is an important direction of textual research and is attracting more and more researchers' attention both at home and abroad.As a semantic resource description and reasoning tool,ontology has been widely applied and studied in many fields of semantic research,such as natural language processing,knowledge management,knowledge engineering,information integration,and so on.Domain ontology is based on the theory of ontology,in which the relationship between concepts and concepts in a particular domain is described in detail.The study of semantic relevance is a research foundation of ontology mapping reasoning,semantic analysis,data analysis and data mining,knowledge management and personalized content recommendation,in which Ontology-based semantic relevance calculation between concepts is a common method.Because the text data is unstructured,the data type is complex and the structure is varied,the computer is difficult to understand and deal with the original text information.The recent study of semantic relevance between texts mainly focuses on the method of calculating the relevance of text by using the concept as the basic processing unit.The accurate and efficient information extraction of the text is often the first step in the calculation of the semantic relevance between the text This paper will use the knowledge points extraction methods,And then the semantic relevance between the text is measured by calculating the semantic relevance between the set of text knowledge points.The work and innovation point of this paper are mainly as follows:(1)The building of the domain ontology of hypertension.In this paper,the domain ontology is built by using Protege ontology editing software and the ontology method called"seven-step method",referring to the "Chinese Hypertension Prevention Guide 2010",,Which provides the data foundation for the subsequent study of semantic relevance.(2)Proposing a comprehensive semantic relevance between concepts calculation method called LCA based on multiple dimensions.By the study of the relevant research theory of semantic relevance at home and abroad,and on the basis of predecessors,this paper points out the problems existing in the semantic correlation calculation method of single dimension,such as the distance-based semantic relevance calculation method,information-based semantic relevance calculation method and feature attribute-based semantic relevance calculation method.A semantic relevance calculation method called LCA is proposed based on this three dimensions.And then the weight of each dimension is estimated is the basis of the domain ontology of hypertension proposed in the paper.Then a contrast experiment will be designed to compare and analyze to prove that the model LCA-KP has higher accuracy in measuring the relevance of text semantics.(3)Proposing a semantic relevance between texts computing model LCA-KP based on ontology theory and knowledge points to calculate the semantic relevance between texts.And then aexperiments will be conducted to prove that the model LCA-KP is valid.This study is a comprehensive study of the method of calculating semantic relevance between texts based on hypertension,and tries to meet the needs of users in text processing in the more professional fields.The text semantic relevance calculation model mentioned in this paper can also be extended to the correlation calculation in other fields.