Improved Genetic Algorithm Based on Variable Weighting FCM Clustering Algorithm

Fuzzy c-means algorithm (FCM) clustering has wide applications, but it is easy to fall into local optimum, and is very sensitive to the initial value. First we determine clustering center of the number and selection of clustering center in fuzzy clustering with genetic algorithm in this paper, introduce the index weight into the FCM and give the iterative formula and the corresponding algorithm. Experimental results show that the method can avoid the defect to a certain extent, in which FCM algorithm is sensitive to the initial value and easy to get into local optimal solution, so that not only the clustering will be more reasonable but also the effect is very good. Springer-Verlag Berlin Heidelberg 2014.