proj10-01

  在试验中编写程序实现了K均值聚类算法,K均值聚类的原理是：在训练样本中找到C个聚类中心，每个聚类中心代表一个类的中心。然后将样本归类到与其最近的聚类中心的那一类。 C的选择是通过先验知识或经验选取的。聚类中心是通过算法迭代求得的。(In the test preparation process to achieve a K means clustering algorithm, K means clustering principle is: in the training samples to find C a cluster center, each cluster center represents a kind of center. Then the samples are classified to the nearest cluster center with the type. C' s choice is a priori knowledge or experience through the selected. Cluster center is obtained through the iterative algorithm.)