Research On Cloud Dimension Optimization Method Of Agricultural Ontology Knowledge

Ontology,the clear and formal standard model of shared conceptualization,becomes the research hot spot of many scholars,and has got a wide-spread application in many aspects,such as knowledge engineering,information processing,natural language understanding,and semantic web.Ontology can realize knowledge sharing and reuse,and make computer understand knowledge in a semantic level;can solve the problem of the semantic heterogeneous in a certain extent.But there is lots of uncertainty knowledge need to express in the process of building ontology.In order to construct a more reasonable,objective and accurate ontology,it is necessary to do research on the theory and methods for dealing with the uncertainty of knowledge.The research of this paper focus on solving problems of the attributes redundancy and the uncertainty representation in agricultural ontology modeling process.Proposed methods of the two key problems,including the cloud dimension optimization of agricultural ontology concept,problem of uncertainty concepts expression in agricultural field.Besides,we put forward the cloud dimension optimization technology and methods for ontology knowledge,and have developed a prototype system of ontology knowledge cloud dimension optimization facing field.The main content and the research achievements are as follows:(1)The cloud dimension optimization model and methods for the concepts in agricultural field concepts is proposed.The model uses clustering algorithm for attribute data category labels,does correlation analysis and redundancy analysis by simplified Filter model to form an ordered sequence.Then uses Wrapper model to detect the classification accuracy of each attribute subset formed by the ordered sequence,choose the best as the cloud dimension of the uncertainty ontology knowledge,and that the model is simple and efficient.The model uses a guideline to optimize the initial clustering center and DB index to determine the K number of cluster,which improves the stability and effectiveness of the model in further.(2)The method of the cloudy format to express uncertainty knowledge is put forward.The method uses cloud transform algorithm to extract agricultural ontology concepts and does ontology concepts merging by the concept merging algorithm to form the hierarchical relationships,realizes the cloudy expression of the uncertainty ontology attributes data.And optimize the steps of calculating En in traditional cloud transform algorithm,which makes the algorithm more efficient and easy to use.Finally to prove effectiveness of the model,we do experiments on the meteorological data of Huangshan tea garden;(3)A prototype system of the cloud dimension optimization of agricultural ontology knowledge is developed.Based on the Matlab platform,a prototype system of cloud dimension optimization of agricultural ontology knowledge is developed,realized the function of cloud dimension optimization and the cloudy expression of the uncertain knowledge.And the validity of the technology and methods is verified through some experiments.The research results has certain theoretical value and practical significance in many aspects,for example,the further study of the theory and method of agricultural cloud ontology modeling,the building of a more accurate and objective agriculture domain knowledge base,and the realization of fully sharing and cooperated service of the agricultural knowledge.