Research And Application Of Variable Precision Multi-granularity Rough Set Model

The variable precision multi-granularity rough set model is an extended model based on the multi-granularity rough set model.The variable precision multi-granularity rough set integrates the error classification rate into the multi-granularity rough set theory to further strengthen the ability of uncertainty classification.In this paper,the importance and reduction of granularity are studied on the basis of variable precision multi-granularity rough set model.The following is the main research work of this paper:(1)In some cases,the traditional importance formula can not distinguish the importance of granularity.In this paper,a new importance measurement method of granularity is proposed based on the variable precision multi-granularity rough set model.In order to further distinguish the importance of different granularity,the reliability coefficient is added to calculate the importance of granularity,which makes the analysis results more reasonable.Finally,the method is validated by a decision information system,and the comparative analysis results show that the method is effective.(2)The granularity matrix suitable for variable precision multi-granularity rough set model is studied.Combined with the granularity matrix,the granularity importance formulas of the variable precision optimistic and pessimistic multi-granularity rough sets were defined,and the algorithm of the variable precision pessimistic multi-granularity rough set reduction was designed based on the granularity importance formula.The results of UCI data set experiments show that the proposed algorithm has higher efficiency than other algorithms.(3)The variable precision variable multi-granularity rough set model adjusts the number of granularity structures satisfying the classification conditions by introducing parameters.The model can further eliminate the limitation of pessimistic and optimistic multi-granularity rough sets with variable accuracy.Some properties of the variable precision variable multi-granularity rough set model are studied.Then,the granularity importance formula of the variable precision variable multi-granularity rough set is defined based on the granularity matrix and the granularity reduction algorithm is designed.The advantages of the proposed algorithm in terms of running time are compared through data set experiments.(4)The reduction algorithms of variable precision pessimistic multi-granularity rough set and variable precision variable multi-granularity rough set are designed.In order to verify the rationality of the algorithm,the reduction algorithm of the two models is applied to the data of a college student’s course score,and the granularity reduction results are briefly analyzed.