Continuous space pattern reduction for genetic clustering algorithm
Chun‐Wei Tsai, Tzu-Yuan Lin, Ming‐Chao Chiang, Chu‐Sing Yang, Tzung‐Pei Hong · 2012
We have recently proposed a highly effective method for speeding up metaheuristics in solving combinatorial optimization problems called pattern reduction (PR). It is, however, limited to problems with solutions that are either binary or integer encoded. In this paper, we proposed a new pattern reduction algorithm named continuous space pattern reduction (CSPR) to overcome this limitation. Simulations show that the proposed algorithm can significantly reduce the computation time of k-means with genetic algorithm (KGA) for solving the data clustering problem using continuous encoding.