Reseraching Of Transfer Learning Methods Based On Optimized Ontology

In recent years, computer science research on the ontology more and more. The so-called ontology, the most famous and widely quoted definition is:Ontology is a conceptual model of a clear specification. Popular speaking, ontology is used to describe an even wider range of areas of the concept and the relationship between the concept of making these concepts and relations in the context of sharing a common recognition, clear, single definition. Thus, machine and machine can communicate. At present, the ontology has been widely used in the semantic web, intelligent information retrieval, information integration, digital library and other fields. Ontology is the core concept of knowledge sharing by reducing the ambiguity of the concepts and terminology, ontology is described as an organization or working group to provide a unified framework or the standard model.In this paper, optimization of ontology around the concept of similarity and the migration of learning t are launched. This paper analyzes the current classical ontology. The ontology generation algorithm is combinated based on Formal Concept Analysis and statistical study, constructing a concept lattice to ontology mapping process. At the same time, this optimization algorithm is used to optimize the ontology. Other, in the ontology mapping process.For ontology mapping conceptual similarity exists in some of the deficiencies.Improved methods have been proposed. Transfer learning is achieved based on an improved method of calculating. This paper describes how to use the ontology approach for knowledge organization with combination of Semantic Web ontology specific characteristics and study for the application entry point vector. Through the above ontological concept similarity and the migration of learning, ontology can be an ideal transfer learning. Ontology knowledge apply to the vector ontology-based knowledge representation and achieve knowledge sharing.