Through principal component analysis, we have synthesized numerous criteria, eliminated information overlapping of the sample, and reduced the input dimension of RBF network.

• 通过主成分分析将众多指标进行综合，消除样本间的信息重叠，降低RBF网络的输入维数。