Research And Application On Logical Disjunct Operator Of Variable Precision And Grade Multi-granularity Rough Set

Multi-Granulation Rough Set(MGRS)is an extended model of classical rough set,which does not require additional prior knowledge.By relying on the characteristics of the data set itself,it can analyze data,discover hidden knowledge and reveal potential rules.MGRS is widely used in index selection and index weight calculation in evaluation system.As an important indicator of air quality monitoring,PM2.5 is closely related to human health and atmospheric environmental quality,which has attracted wide attention of scholars.It is particularly important to study appropriate models to analyze the influencing factors of PM2.5.In real life,the final meteorological information is often incomplete because of the errors of measurement data and the limitations of data acquisition.In order to obtain more accurate knowledge from incomplete information systems,this paper devotes to building a new type of incomplete MGRS.Firstly,in view of the shortcomings of the existing MGRS,logical disjunct operator of variable precision and grade MGRS involved limited tolerance relation is proposed,which includes optimistic and pessimistic forms.Secondly,by analyzing the relaxation and strictness of lower approximation conditions under the logical disjunct operator of variable precision and grade optimistic MGRS and the logical disjunct operator of variable precision and grade pessimistic MGRS,the variable MGRS based on logical disjunct operator of variable precision and grade is presented,the measure relations of these kinds of rough sets are discussed.Finally,according to the characteristics of PM2.5meteorological data,the relationship between PM2.5 and its influencing factors is briefly analyzed.Data preprocessing is carried out through a series of methods such as continuous value discretization and attribute value coding,and the proposed new MGRS is reasonably used in PM2.5 meteorological data.The feasibility of the MGRS model is verified by comparison test and result analysis.The proposed MGRS model is used to calculate the upper and lower approximation of PM2.5 data,which lays a foundation for the weight analysis of PM2.5 influencing factors in the future.