CA算法可以将数量型属性划分成若干个优化的区间，它综合了分层聚类于划分聚类的优点，对于给定的不同的初始类个数，CA算法能随着迭代过程的不断进展改变类的数目，一些竞争力差的类即类的基数小于给定阙值的类将在迭代过程中不断消失，最终得到能够有效体现数据实际分布情况的优化聚类个数。-CA algorithm for quantitative attributes can be divided into a number of the interval optimization, hierarchical clustering which combines the advantages of clustering in the division, for a given number of different types of initial, CA algorithm as the iterative process the continuous progress of change in the number of categories, some categories of poor competitiveness of the base type that is less than the value of a given category Que iterative process will continue to disappear and eventually be able to effectively reflect the actual distribution of data to optimize the number of clustering.