论文部分内容阅读
讨论的是对模糊C-均值聚类方法的改进,在原有的模糊C-均值算法的基础上,提出一种软硬结合的快速模糊C-均值聚类算法。快速模糊C-均值聚类算法是在模糊C-均值聚类算法之前加入一层硬C-均值聚类算法。硬聚类算法能比模糊聚类算法以高得多的速度完成,将硬聚类中心作为模糊聚类中心的迭代初值,从而提高模糊C-均值聚类算法的收敛速度,这对于大量数据的聚类是很有意义的。用数据仿真验证了这种快速模糊C-均值聚类算法比模糊C-均值算法迭代调整过程短,收敛速度快,聚类效果好。
The improvement of fuzzy C-means clustering method is discussed. Based on the original fuzzy C-means algorithm, a fast and soft C-means clustering algorithm based on hardware and software is proposed. The fast fuzzy C-means clustering algorithm adds a layer of hard C-means clustering algorithm before the fuzzy C-means clustering algorithm. The hard clustering algorithm can be completed at much higher speed than the fuzzy clustering algorithm, and the hard clustering center is used as the iterative initial value of the fuzzy clustering center so as to improve the convergence speed of the fuzzy C-means clustering algorithm, The clustering is very meaningful. The data simulation shows that the fast fuzzy C-means clustering algorithm has a faster convergence rate and better clustering effect than the fuzzy C-means algorithm.