Research on Attribute Reduction Algorithm of 0-1 Decision Information Table

Abstract

In order to study the attribute reduction of the decision information table, three different modes are adopted. The first mode performs attribute reduction under the premise that different attributes have the same weight, and performs attribute reduction on the given table by using a blindly deleted attribute reduction algorithm; the second mode is under the premise of considering different weights, Carry out attribute reduction, establish a mathematical model about the discernibility matrix through the information table, and perform attribute reduction on the information table according to the attribute reduction algorithm of the Skowron discernibility matrix; The third is to first calculate the total value of the attribute through the decision information table, and then consider the influence of each attribute on the total value of the attribute, establish a mathematical model according to the attribute reduction algorithm of Pawlak attribute importance, and perform attribute reduction according to the attribute importance. Theoretical analysis and simulation results show that the algorithm proposed in this paper is effective and feasible, improves the efficiency of attribute reduction, and provides a certain theoretical basis for attribute reduction.