An Algorithm of Attribution Reduction of Variable Precision Rough Sets Theory

Attribute reduction is one of the key topics in the Rough Set theory field.It has been proved that computing the minimal reduction of decision table is an NP-hard problem.Relative discernibility matrix and attribute significance are considered to solve the attribute reduction of variable precision rough sets theory.Moreover,the improved algorithm of attribution reduction reduction is given.It can simplify the operation and enhance reduction the efficiency of seeking the reduction in some extent.At last,a practical example is given to show the validity of the algorithm.