BIRCH(Balanced Iterative Reducing and Clustering using Hierarchies)是一个综合的层次聚类算法。它用到了聚类特征(Clustering Feature, CF)和聚类特征树(CF Tree)两个概念，用于概括聚类描述。聚类特征树概括了聚类的有用信息，并且占用空间较元数据集合小得多，可以存放在内存中，从而可以提高算法在大型数据集合上的聚类速度及可伸缩性。 -BIRCH (Balanced Iterative Reducing and Clustering, using Hierarchies) is a comprehensive hierarchical clustering algorithm. It uses the characteristics of clustering (Clustering, Feature CF), and the two concepts of clustering feature tree (CF Tree), used to summarize the clustering descr iption. The clustering feature tree summarizes the useful information of the clustering and space than a much smaller data sets can be stored in memory, which can improve the clustering algorithm on large data sets, speed and scalability.