针对模糊C-均值(FCM)聚类算法的容易收敛于局部极值的不足,提出了一种改进的模糊FCM聚类算法,此新算法在聚类中心选取和优化过程中进行了充分的考虑,是一种用于确定最佳聚类数的聚类算法,并且利用了分阶段思想,结合动态直接聚类算法和标准聚类算法,来尽量避免模糊C-均值(FCM)聚类算法的不足。新算法与传统(FCM)聚类算法方法相比,提高了算法的寻优能力,并且迭代次数更少,在准确度上也有较大的提高,具有很好的实际应用价值。
For the fuzzy C-means (FCM) clustering algorithm of the shortcomings, an improved fuzzy clustering algo- rithm FCM is proposed. The fuzzy C-means (FCM) clustering algorithm is easy to converge to local extremurn. The new al- gorithm has taken into account in the algorithm selection and optimization of the process of cluster centers. It is a new algo- rithm that is used to determine the optimal number of clusters of the clustering. It uses a phased thinking and dynamic cluste- ring algorithm and standard clustering algorithm to avoid the deficiencies of the ambiguity C-Means (FCM) clustering algo- rithrru Compared with the traditional (FCM) clustering algorithm the new algorithm improves the optimization ability of the algorithm, and the number of iterations is fewer. In addition, there is also a larger improved at the accuracy, and good practical applications.