应用新的模式识别方法PCA－BPN（PrincipalComponentAnalysis－BackPropagationNetwork）指认CmⅠ奇宇称未知能级，支持了前人应用传统的KNN（KNearestNeighbors）等模式识别方法及对传神经网络方法（CounterPropagationNetwork，CPN）对大部分谱线的指认，进一步确认了这些组态的归属；鉴别了KNN等与CPN不同的预报结果，纠正CPN的某些错误分类，并以可视非线性映照分类器加以佐证 Applying the new pattern recognition method PCA-BPN (Principal Component Analysis-Back Propagation Network) to identify Cm I odd-parity unknown level, it supports the predecessors to apply the traditional pattern recognition methods such as KNN (KNearest Neighbors) and CPN (Counter Propagation Network) The identification of most of the spectral lines further confirmed the ownership of these configurations; identified different forecasting results, such as KNN, from CPN, corrected some misclassification of CPN, and corroborated by visual non-linear mapping classifier