Fuzzy Ontology Extension And Induction Based On Cognitive Model

Ontology is an important tool for expressing semantics because it can clearly express the features of concepts and relationships between concepts,and is the core technology of the Semantic Web.However,most of the concepts acquired in natural language are often ambiguous,but traditional ontology is not enough to represent such information.Therefore,fuzzy ontology is used to represent fuzzy concepts and to make fuzzy inference.Because the representation of knowledge in fuzzy ontologies is very close to the form of human cognitive concepts,in cognitive computing,fuzzy ontology is widely used.Cognitive computing is derived from the artificial intelligence of a computer system that mimics the human brain.Its purpose is to enable computers to simulate the process of autonomous learning of humans in the cognitive world.With the gradual expansion of fuzzy ontology in the real world,the requirements for the expansion and accuracy of the automatically constructed fuzzy ontology will also increase.The fuzzy ontology should be able to automatically expand and update itself,completing the replacement of old and new knowledge,and make the existing fuzzy ontology more refined and accurate.Fuzzy ontology learning can make fuzzy ontology complete its own knowledge from scratch,and optimize the process of self-learning of existing knowledge.Cognitivebased fuzzy ontology learning,based on cognitive computing,it proposes the cognitive strategies extension,reduction,induction,update and revision for fuzzy ontology learning.Using these cognitive strategies,fuzzy ontology can simulate the human's ability to learn and optimize knowledge,enabling computers to interact with humans more smoothly.This paper proposes a cognitive-based fuzzy ontology extension and induction method.It focuses on the two strategies of extension and induction in the cognitive model of fuzzy ontology learning which simulates the inductive learning process of the human cognitive world and can be used for automatic construction of fuzzy ontology and subsequent knowledge extension and structural optimization.The dissertation combines cognition calculations to define a cognitive model of fuzzy ontology learning.The structured knowledge base Hownet is used as a starting point for learning.For this knowledge system,this paper gives the fuzzy representation of concepts and relationships and the calculation of fuzzy membership degree.The corresponding cognitive strategies are proposed on the cognitive model of fuzzy ontology learning.When new knowledge is added to the existing fuzzy ontology,using Fuzzy C-Means algorithm to cluster new knowledge,then the new knowledge can be added by the axioms and operators proposed by the extension strategy.The shared knowledge at all levels of the fuzzy ontology realizes the refinement of knowledge through the principles and operators of the fuzzy ontology induction strategy,which makes the fuzzy ontology more generalizable,and reduces redundancy effectively,further enhancing the selflearning ability of the fuzzy ontology.Finally,a part of the fuzzy ontology is presented to show the process of extension and induction of the fuzzy ontology and the result,and the comparison between the extended inductive operation of the fuzzy ontology and the traditional ontology is compared.Experiments show that the extension and induction of cognitive-based fuzzy ontology is much better than traditional ontology in accuracy and efficiency,and it performs well in the automatic construction and self-improvement of fuzzy ontology.