Algorithm Design of Navigation Satellite Evaluation Classification Rules Based on VPRS and DPSO

In order to make the classification rules more reliable in dealing with the decision data of evaluation, a rough rule mining approach is proposed. With uncertain measurement of rough set, lower approximation reduction based on variable precision rough set (VPRS) is analyzed. The definition of measure in decision table is given based on information entropy measure. Then based on discrete particle swarm optimization algorithm (DPSO), an algorithm for knowledge reduction in variable rough set is presented. The experimental result shows that this algorithm has a certain classification capability for the decision data, and the rules obtained by this approach have more precision and better overlay.