摘要
为了克服传统层次聚类算法由于两类合并造成的中心点偏移的严重缺陷,提出了一种基于类中心矫正的层次聚类算法,从而提高了算法的精确度;同时继承了传统层次聚类对初始中心点的无依赖性;经分析,算法对于已知聚类数和未知聚类数两种情况均有着良好的聚类效果.通过标准数据测试,结果表明新算法的聚类性能与层次聚类算法相比有更高的精确度;并且让新算法用于指导图像分割实验,证明了算法的有效性.
In order to solve the problem of class center offset caused by merger of class of traditional hierarchical clustering algorithm,an algorithm of hierarchical clustering is proposed based on correcting class center.The accuracy of proposed algorithm is improved.And it inherits the non-dependnce of the initial clustering center of the traditional hierarchical algorithm.After analysis,the algorithm obtains a good clustering result for the known and unknown number of clusters.It is proved that the proposed algorithm is not only more precise than traditional hierarchical clustering algorithm,but also the clustering results show that the algorithm is effective in image segmentation.
出处
《微电子学与计算机》
CSCD
北大核心
2011年第10期217-220,共4页
Microelectronics & Computer
基金
江苏省自然科学基金项目(BK2009393)
南京林业大学基金项目(163070053
163070657)
关键词
层次聚类
偏移
类中心矫正
精确度
hierarchical clustering
offset
class center correction
accuracy