Description: K-MEANS算法
输入：聚类个数k，以及包含 n个数据对象的数据库。
输出：满足方差最小标准的k个聚类。
处理流程：
（1） 从 n个数据对象任意选择 k 个对象作为初始聚类中心；
（2） 循环（3）到（4）直到每个聚类不再发生变化为止
（3） 根据每个聚类对象的均值（中心对象），计算每个对象与这些中心对象的距离；并根据最小距离重新对相应对象进行划分；
（4） 重新计算每个（有变化）聚类的均值（中心对象）-K-MEANS algorithm Input: cluster number k, and contains n data object database. Output: the minimum standards to meet the variance k-clustering. Deal flow: (1) a data object from the n choose k object as initial cluster centers (2) cycle (3) to (4) until a change in each cluster is no longer so far (3) according to each Clustering objects mean (central object), calculated for each object with these centers to object distance and in accordance with a minimum distance between a re-division of the corresponding object (4) re-calculated for each (change) clustering of the mean (central object ) Platform: |
Size: 3072 |
Author:快快 |
Hits:
Description: 改进的k-means方法，对聚类的实例节能型加权 少数类多数类的函数-Improved k-means method for clustering a small number of examples of energy-saving type of weighted majority of types of function Platform: |
Size: 9216 |
Author:zhangwen0927 |
Hits:
Description: 文档分类，用K均值，加入了K的选择算法，不用人为设定聚类个数-Document classification, using K-means, joined the K of the selection algorithm, not the number of artificial clustering Platform: |
Size: 20480 |
Author:王 |
Hits:
Search - k-mean number clustering in java