Relative Reduction Of Neighborhood-covering Multigranulation Rough Set Based On Evidence Theory

With the development of social science and technology,the scale of data collected by human beings is getting larger and larger,and the data structure is becoming more and more complex.How to acquireeffective knowledge from large and complex data has become a common demand in many fields.As an important mathematical tool of granular computing,rough set provides an effective method of knowledge discovery and knowledge reduction,which has been widely used in the field of artificial intelligence,pattern recognition and machine learning.As the extension model of classical rough set,covering rough set and multigranulation rough set are more suitable for practical application in real life,which has attracted the attention of many scholars.This paper combines the model of covering rough set and multigranulation rough set,and takes the model of neighborhood-covering multigranulation rough set as the main research objects.The research contents include the relationship between neighborhood-covering multigranulation rough set and evidence theory,the belief structure of neighborhood-covering multigranulation rough set based on covering decision,the relative reduction of neighborhood-covering multigranulation rough set and the reduction of neighborhood-covering multigranulation rough set based on covering decision.The main contributions of this paper are listed as follows:(1)The belief structureof neighborhood-covering multigranulation rough setsbased on covering decision and partition decision are studied.In this paper,the belief functions and plausibility functionsfrom evidence theory are employed to characterize the approximations of neighborhood-covering multigranulation rough set based on covering decision and partition decision.(2)The relative reductionof neighborhood-covering multigranulation rough set based on covering decision and partition decisionare studied.In this paper,the algorithms for computing relative reduction of neighborhood-covering multigranulation rough set based on covering decision and partition decision are proposed according to the significance of coveringsdefined by the belief function and plausibility function,and its rationality are examined by a practical example.