Table 2 Comparison of the classification effects of various data combinations.

From: Integrating Internet multisource big data to predict the occurrence and development of COVID-19 cryptic transmission

Model

Dataset

PRE

SEN

SPE

ACC

F1

ROC-AUC

PR-AUC

USEE1

Training

0.6588

0.9825

0.8535

0.8824

0.7887

0.9791

0.9112

 

Testing

0.6364

0.7368

0.8788

0.8471

0.6829

0.8086

0.6103

USEE2

Training

0.7368

0.9825

0.8990

0.9177

0.8421

0.9890

0.9303

 

Testing

0.6667

0.8421

0.8788

0.8706

0.7442

0.8445

0.7846

USEE3

Training

0.8594

0.9649

0.9546

0.9569

0.9091

0.9908

0.9480

 

Testing

0.7619

0.8421

0.9242

0.9059

0.8000

0.9553

0.8327

  1. USEE under-sampling synchronous evolutionary ensemble. PRE precision, SEN sensitivity, SPE specificity, ACC accuracy, F1 F1-Score, ROC-AUC receiver operating characteristic-area under curve, PR-AUC precision recall-area under curve.
Back to article page