Research On The Construction Of Fuzzy Ontology Based On Fuzzy Formal Concept Analysis And Fuzzy Concept Similarity

Semantic Web is the expansion and extension of the current World Wide Web,and its purpose is to make the network is not just a display of the information platform,but also by the computer to "understand" and automatic processing, betterhuman-computer interaction. Ontology is the foundation of the semantic web, providethe conceptual domain common recognition and its relationship to the Semantic Web.Along with the depth study of Ontology, people realize that it is difficult to express the uncertaininformation in the real world using the Classical Ontology. Therefore research on the building offuzzy ontology become the hot spot of the Semantic Web.The manual building of fuzzy ontology is time-consuming and laborious, soautomation building method become the mainstream of current research. Theapplication of fuzzy formal concept analysis can automatically constructed fuzzyconcept lattice by uncertain information, which can be used as the basis ofconstructing fuzzy ontology. But the fuzzy concept lattice has the phenomenon ofredundancy concept, which need clustering in order to achieve the purpose of simplifythe quantity of the fuzzy concept and the size of concept hierarchy. During the existingclustering methods, fuzzy formal concept similarity calculation still haveshortcomings, such as do not take into account the affect of the fuzzy concept’sextension’s similarity and the intent’s same attributes in the clustering, ultimatelyaffect the quality of the generated fuzzy ontology. Based on this, this paper presents aconstruction method of fuzzy ontology based on fuzzy formal concept analysis andfuzzy concept similarity. First extract uncertain information in the domain ontologyconstructed single-valued fuzzy formal context. Use the incremental algorithm totransform the single valued formal context into fuzzy concept lattice. Then, using theclustering algorithm based on fuzzy concept similarity, calculate the similarity offuzzy form concept of extension and intension, and the weighted factor is introduced to adjust the extension and intension of similarity ratio, adds the concept clusteringcompleteness, accuracy and flexibility. Finally the use of the clustering algorithm offuzzy concept lattice, generating fuzzy concept clustering nodes and their hierarchicalrelationships, and put forward a mapping rule to fuzzy ontology mapping. Finally, thesimulation experiments and practical examples of the proposed algorithm, the resultsshow that the proposed method improves the clustering quality of fuzzy concept, aswell as flexibility.