特征组

模型

AUC

准确率

灵敏度

特异度

NO.1

KNN

0.635

0.717

0.855

0.416

SVM

0.872

0.902

0.952

0.792

RF

0.94

0.959

0.99

0.89

LightGBM

0.956

0.966

0.982

0.929

NO.2

KNN

0.669

0.725

0.82

0.518

SVM

0.653

0.759

0.937

0.369

RF

0.869

0.895

0.94

0.797

LightGBM

0.857

0.885

0.932

0.781

NO.3

KNN

0.613

0.679

0.792

0.434

SVM

0.59

0.716

0.93

0.251

RF

0.798

0.845

0.925

0.671

LightGBM

0.817

0.849

0.902

0.732

NO.4

KNN

0.608

0.682

0.807

0.409

SVM

0.514

0.681

0.962

0.066

RF

0.72

0.787

0.9

0.539

LightGBM

0.742

0.8

0.9

0.583

NO.5

KNN

0.526

0.634

0.817

0.235

SVM

0.5

0.682

0.99

0.011

RF

0.63

0.728

0.894

0.365

LightGBM

0.653

0.73

0.862

0.444

NO.6

KNN

0.709

0.751

0.822

0.596

SVM

0.759

0.819

0.922

0.595

RF

0.867

0.899

0.952

0.781

LightGBM

0.898

0.921

0.96

0.836

NO.7

KNN

0.717

0.765

0.845

0.589

SVM

0.832

0.866

0.925

0.738

RF

0.878

0.912

0.97

0.786

LightGBM

0.905

0.926

0.962

0.847

NO.8

KNN

0.665

0.723

0.822

0.508

SVM

0.691

0.78

0.93

0.452

RF

0.911

0.933

0.97

0.852

LightGBM

0.961

0.971

0.988

0.934

NO.9

KNN

0.709

0.751

0.822

0.596

SVM

0.782

0.828

0.905

0.659

RF

0.901

0.928

0.972

0.83

LightGBM

0.955

0.969

0.992

0.918

NO.10

KNN

0.717

0.765

0.845

0.589

SVM

0.793

0.843

0.927

0.659

RF

0.875

0.911

0.97

0.78

LightGBM

0.959

0.972

0.995

0.923

NO.11

KNN

0.726

0.768

0.84

0.612

SVM

0.797

0.839

0.91

0.685

RF

0.89

0.921

0.972

0.808

LightGBM

0.961

0.973

0.992

0.929

NO.12

KNN

0.635

0.72

0.865

0.405

SVM

0.855

0.89

0.95

0.759

RF

0.916

0.945

0.992

0.84

LightGBM

0.964

0.973

0.988

0.94