An improved genetic FCM clustering algorithm

In dealing with such defects of the genetic FCM (Fuzzy C-Means) clustering algorithm as long calculating time and poor clustering results, this paper proposes an improved algorithm which improves the crossover, selection, and mutation parts of the GA (genetic algorithm), enhances its global searching capability and eases the difficulty in setting up genetic parameters. At the same time, this improved algorithm performs FCM optimization immediately after each generation of genetic operation, which increases the converging speed. As the experiment shows, the clustering results, converging speed and stability of the improved algorithm are much better than the algorithms in references.