Attribute Reduction Algorithm Research Based on Golden Section and Back Elimination

Data mining and analysis algorithms are known to degrade in performance when facing with many redundant or irrelevant features. Attribute reduction is one of the primary problems of rough set theory, the goal of which is to delete irrelevant or unimportant information. Once all attribute reducts are got, the reasoning capability with multi attributes absent can behave well. Thus how to get all attribute reducts is worth a problem to research. In this paper, an algorithm based on golden section and back elimination is presented for getting all attribute reducts of decision system. Experiment results show the validity of our proposed algorithm.