Attribute reduction algorithm of rough set fused with ant colony algorithm

摘要

This paper fused rough set theory and ant colony algorithm. The attribute core was determined through the correlative algorithms of rough set, which could be used as initial node of ant colony algorithm, then the time complexity and search space were reduced. The search capacity was used to get the lease combination of these nodes, which is the minimal attribute set, and the NP-hard problem in attribute reduction by using rough set was avoided. The result of experiment showed the feasibility and validity of this algorithm.