摘要
在典型的空间聚类算法K-平均法和K-中心法中,K一般为用户事先确定的值,然而,实际中K值很难被精确地确定,往往表现为一个模糊的取值区间。在此提出距离代价函数的概念,建立了相应的数学模型并设计了一个新的K值优化算法,对空间聚类K值优化问题进行了初步的研究。
The value of K is always confirmed in advance to exert K-means algorithm of spatial clustering. However, it can not be clearly and easily confirmed in fact for its uncertainty. A distance cost function was recommended. A corresponding math model was set up and a new optimization algorithm of K value was designed. A preliminary study on the optimization of K value for spatial clustering was realized by a simulation design.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2006年第3期573-576,共4页
Journal of System Simulation
基金
国家自然科学基金(70471046)
关键词
空间聚类
尽平均算法
距离代价函数
K值优化
spatial clustering
K-means algorithm
distance cost function
optimization of K.
