感谢楼主,请问楼主,使用kmeans进行聚类时,如何查看最后聚类模型的效果呢??
算法如下:
proc fastclus data=ketizu.Sw2014combine1_shouji2 summary maxc=4 maxiter=99
outseed=clusterseed out=ketizu.Sw2014combine1_shouji3 cluster=cluster least=2;
id yhhm;
var yuedu_flow yuedu_PV gouwu_flow gouwu_PV youxi_flow youxi_PV yinyue_flow ;
run;
输出结果如下:
Criterion Based on Final Seeds = 0.8741
Cluster Summary
Maximum Distance
RMS Std from Seed Radius Nearest Distance Between
Cluster Frequency Deviation to Observation Exceeded Cluster Cluster Centroids
--------------------------------------------------------------------------------------------------
1 572157 0.5568 11.9042 2 3.3507
2 308068 1.0661 17.2514 1 3.3507
3 98032 1.4481 19.3305 2 4.6291
4 29549 1.1188 17.1095 2 7.5117
Pseudo F Statistic = 103722.1
Observed Over-All R-Squared = 0.23592
Approximate Expected Over-All R-Squared = 0.06032
Cubic Clustering Criterion = 7249.208
WARNING: The two values above are invalid for correlated variables.