Abstract:

The research of uncertainty measure is the key point of the rough sets’ theory. The uncertainty of decision information system is produced by boundary region. Based on the theory of variable precision rough sets, the objects of boundary region are divided into weak-consistent objects and inconsistent objects. This paper builds an uncertainty measure criterion on conditional entropy of boundary region. Theoretical analysis proves that in case of weighing the consistency of decision system, this conditional entropy of boundary region have something in common on recognition with the normal conditional entropy, and this measure criterion is also equivalent to the relative regular domain in feature reduction of rough sets.