摘要
RNN(相互最近邻)算法是一种基于层次的聚类算法,它比其他传统的层次聚类算法聚类更快.由于利用RNN算法对同一个数据集聚类,若选择不同簇间距离度量方式,那么聚类结果就会不同.因此在分析聚类结果对距离度量方式依赖性的基础上,采用用聚类聚集的思想,找出一种新的聚类方式,从而使得聚类效果更好.
RNN （reciprocal nearest neighbor）algorithm is a kind of clustering algorithm which is based on hierarchy. Compared with other traditional hierarchical clustering algorithm, RNN algorithm is much faster. Using the same data set by RNN algorithm clustering, if we choose different measure way between cluster distance, then the clustering results will be differ- ent. By analyzing the clustering results from the dependence of the way to measure, on this basis, we can gure out a new clustering way which made the clustering effect better.
出处
《太原师范学院学报（自然科学版）》
2012年第2期72-75,共4页
Journal of Taiyuan Normal University:Natural Science Edition
关键词
聚类聚集
最近邻
聚类
RNN
clustering aggregation
nearest neighbor
clustering
RNN
