Research On Knowledge Reasoning Supported By Semantic Web Rough Fuzzy Ontology

As an extension of the semantic Web fuzzy ontology, the semantic Web rough fuzzy ontology can also be regarded as an extension of the semantic Web ontology by a first fuzzy-second rough method, which is composed by rough fuzzy concepts and relationships between them. As a representation form of imprecise knowledge, the semantic Web rough fuzzy ontology expresses not only fuzzy knowledge but also rough knowledge to reflect the human’s imprecise cognition on the objective world.Knowledge reasoning is the basis of knowledge application. The nature of knowledge is imprecise. With the further research on imprecise knowledge, more and more researchers are paying their attention to imprecise knowledge reasoning.In this thesis, the semantic Web rough fuzzy ontology is applied to knowledge reasoning; then, a method of knowledge reasoning based on the rough fuzzy ontology is designed; next, a knowledge reasoning framework supported by the semantic Web rough fuzzy ontology is proposed, at last, the knowledge reasoning for imprecise knowledge is achieved.Firstly, the model, construction, expression, and application of the semantic Web rough fuzzy ontology are discussed. Based on the researches on the semantic Web fuzzy ontology and the semantic Web rough ontology, the definition and the model of the semantic Web rough fuzzy ontology are proposed in this thesis, and a construction method based on concept lattice and an expression method based on OWL are discussed. After these works above, the application of the semantic Web rough fuzzy ontology in the field of knowledge acquisition, knowledge reasoning and. information semantic retrieval are investigated.Secondly, the rules implied in the semantic Web rough fuzzy ontology are extended. By analyzing the relationships between class and class, class and instance, instance and properties, et al, and the rules of the semantic Web rough fuzzy ontology are extended by SWRL.Thirdly, the semantic Web rough fuzzy ontology and the SWRL rules are translated into the Prolog knowledge base. XSLT is used to translate the semantic Web rough fuzzy ontology into a Prolog rough fuzzy fact base, and translate the SWRL rules into a Prolog rough fuzzy rule base. The Prolog rough fuzzy fact base and the Prolog rough fuzzy rule base are used to constitute the knowledge base in the knowledge reasoning system.Finally, the knowledge reasoning system supported by the semantic Web rough fuzzy ontology is designed and developed. To implement the system, the rough fuzzy knowledge matching method and the rule conflict resolution strategy are designed. Depending on these, the knowledge reasoning system supported by the semantic Web rough fuzzy ontology is implemented, and the experimental data are used to assess the performance of the system. The experiment result shows that the accuracy rate of the system is better than other’s.