Granule Description Of Object-Oriented(Attribute-Oriented)Concept Based On GrC And FCA

In recent years,with the rapid development of science and technology,people's demand for data is increasing,and data mining has become a popular field.As the main tools for knowledge acquisition,formal concept analysis and granular computing theory have been widely used in data mining,knowledge discovery,artificial intelligence,rule extraction,information retrieval and other fields.Formal concept analysis theory is a primary tool for knowledge representation and knowledge discovery,and describing the generalized and instantiated relations between formal concepts.As a rapidly developing theoretical discipline in recent years,Granular computing theory has become an indispensable theory in the field of artificial intelligence,which is favored by researchers.Granular computing theory is a method of solving complex problems by decomposing and transforming complex problems through the process of information granulation to obtain multiple relatively simple subproblems,and then solving the granulated subproblems one after another.The basis of granular computing is information granulation,that is,under certain criteria,multiple information granules are generated by dividing the problem space,and these granules are used to describe some of the information contained in the problem itself.The partial order relationship between these grains forms a multi-level grain structure.From the perspective of formal concept analysis and granular computing,this paper focuses on the granular description of object-oriented(attribute-oriented)concepts based on existing theories,and the main work and research result are as follows:(1)Using the existing theories,the granular description of the object-oriented(attributeoriented)concept is studied.First,the object granules and attribute granules are constructed by defining the unary and binary object(attribute)descriptors,and described separately.Then,by comparing the accuracy of the three approximate descriptions,the feasibility and effectiveness of the description method are verified;(2)In formal decision context,combining the classical formal background and the theoretical basis of granular computing,the granule description based on attribute granular coordination formal decision context is discussed,and consistent with the granular description of the objectoriented(attribute-oriented)concept,the unary and binary object(attribute)descriptors of the object-oriented(attribute-oriented)concept based on attribute granular coordination formal decision context are discussed,and then,The decision rules are expanded by the approximate description of undefinable object(attribute)descriptors.Granular computing is one of the most effective methods for solving problems in the current data mining direction.The combination of granular computing and formal concept analysis theory makes the process of knowledge discovery more detailed,which can better enable readers to learn and understand the process of knowledge discovery.