The purpose of this article was to introduce the basic concepts, computational methods, two examples and their implementation with SAS related to cluster analysis of disordered samples. The basic concepts included entropy, information entropy, synchronization and normalization, entropy value and diversity coefficient, weight coefficient. The calculation method involved the principle of calculation and the formula of entropy value method. The data in the two examples were respectively "the survey results of inpatient healthcare quality indicators of Zhongda Hospital Affiliated to Southeast University from 1990 to 1999" and "the survey results of independent innovation ability indicators of 15 high-tech parks". Using SAS software, the data in the two examples were analyzed by cluster analysis of disordered samples, and the results of classification were given based on the optimal segmentation method, and the output results of SAS were explained. |