数学建模评价模型中常用到的聚类分析法，聚类是将一组给定的未知类标号的样本分成内在的多个类别，使得同一类中的样本具有较高的相似度，而不同类中的样本差别大，聚类分析的目的是揭示和刻画数据的内在结构。(The clustering analysis method commonly used in mathematical modeling evaluation model. Clustering is to divide a set of samples of unknown class marks into multiple categories, which makes the samples of the same class have a high similarity, and the differences in the different samples are large. The purpose of clustering analysis is to reveal and describe the internal structure of the data.)