摘要
为了得到最佳聚类数和相应的每一类中的样本 ,文中首先介绍了一种新聚类方法 ,用该方法构造了一个既考虑类与类之间的分散程度、又考虑同一类紧凑程度的目标评价函数 ;再运用模糊c -均值算法 (FCM )进行迭代 ,求得每一类的中心和隶属度值 ;然后运用遗传算法搜索全局极值点 ;最后运用该算法对我国全要素生产力进行了模糊分类 .
In order to get the optimal clustering number and the corresponding samples in each cluster, a new clustering method is introduced in this paper. The new method can be used to construct an object evaluation function,which takes into consideration not only the scattering degree among clusters but also the compactness in the same cluster. To obtain the center and memberships values of each cluster by iteration, the FCM(Fuzzy c-mean) algorithm is adopted. The genetic algorithm is then used to search for the global optimal point. The proposed algorithm is applied to the fuzzy clustering of the full-factored productivity of China.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2004年第10期93-96,共4页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金资助项目 (70 2 730 4 4 )
