Skip to Main Content
The discretization of continuous attributes is one of the main problems in rough sets theory. Because there are no uniform criterion for the selection of the cut-points set, the character of mathematic curve is considered and a new algorithm of continuous attribute discretization based on optimization is presented. In this algorithm, the set of candidate cut-points is obtained according to the property of curve inflection points, moreover, quantum particle swarm optimization (QPSO) algorithm is used to search and optimize the set of candidate cut-points. The continuous attribute discretization is complete and the experiment shows that the algorithm is feasible and effective.
Date of Conference: 25-27 June 2008