Research On The Methods Of Adaptive Learning For Tea Domain Ontology

Ontology learning is a series of methods and technologies of constructing ontologyautomatically or semi-automatically, which greatly promotes the research of ontologymodeling, mainly reflected in not only reducing manpower on the complex hand-builtontology also improving ontology construction quality. However, in our process ofontology modeling and ontology studying and researching in the field of agriculture,information in the tea domain knowledge corpus is found to remain increasing andchanging continuously. In order to adjust the ontology established to these changes on itsown, it is necessary and important to do research on the learning theory and methods ofontology adaptive learning in agricultural domain.Based on the existing research achievement, and concentrating on the tea domainadaptive problems on ontology learning, this paper studies with emphasis the method of teadomain ontology adaptive learning, mainly including adaptive extraction of tea domainconcepts, adaptive extraction of tea domain relations between concepts and the integratedmodel design of adaptive ontology learning model system of tea domain. The maincontents and achieved results are as follows.1. This paper proposes adaptively extracting the tea domain concepts. The proposedmethod constructs a model of Bayesian network combining context dependency analysis,domain dependencies and mutual information. By Bayesian being reasoned backward,does it eventually achieves adaptive extraction of the tea domain concepts.2. This paper proposes the theory and method of the tea domain relations betweenconcepts adaptive extraction. By means of the taxonomic relations between the tea domainconcepts by conceptual clustering and clustering, and by Bayesian being reasonedbackward, does it eventually achieves adaptive extraction of the relations between teadomain concepts.3. A tea domain ontology adaptive learning system is developed. Adopting Servlet/Jsp programming technology, a tea domain ontology adaptive learning system is developedin Eclipse platform. We can verify the validity of the theories and methods in this paper byexperiment.For further study of the theory and method of agricultural domain ontology,structuring large-scale agricultural ontology, further establishing the agricultural semanticnetwork and knowledge grid, and realizing agricultural knowledge full sharing and cooperative service, the research findings of this paper have the oretical research value andpractical significance.