Research On Dynamic Approximate Updating And Multi-granularity Modeling Of Intuitionistic Fuzzy Information System

At present,massive data information plays an important role in every intelligent industry and business field.It is an essential problem that how to choose the appropriate data mining method to obtain effective information from the constantly updated data.As we all know,the rough set theory can deal with the update of dynamic data in time,and it is a very effective data mining method.In the information system,the granular structure of the system will change when redundant attributes are deleted or new attributes are added,which will lead to the change of the two approximate operators.In the intuitionistic fuzzy information system,the kernel similarity relationship is established based on the kernel function and the corresponding rough set model is also constructed in this thesis.Besides,the rough approximation update is also studied in the case of dynamically changing attributes.Furthermore,a multi-granularity rough set model based on the dominance relation is established in the interval-valued intuitionistic fuzzy information system,which lays a good theoretical and application foundation for effectively dealing with uncertainty and fuzziness.The specific research contents are as follows:1.In the intuitionistic fuzzy information system,the kernel function is used to measure the similarity between different objects,and the kernel similarity relationship is established according to the similarity.Based on this,the corresponding rough set model is constructed and we also study the basic properties about the model.2.In the case that the attributes set changes with time in the intuitionistic fuzzy information system,in ordered to realize the update of rough approximations quickly,two dynamic update methods are proposed in the thesis.Rough approximation is updated by deleting and adding attributes.In particular,two corresponding dynamic updating algorithms are designed,and the effectiveness of the proposed update method is evaluated through experiments.3.In the interval-valued intuitionistic fuzzy information system,the sorting rules between objects are formulated,and the corresponding rough set models are established accordingly.Based on logical operations,approximate operators with different granularities are recombined to establish two different multi-granularity rough sets.On this basis,a generalized multi-granularity rough set model is further proposed between the two multi-granularity rough set models,which further increases the flexibility and problem-solving ability of rough set theory and fuzzy set theory.