Skip to main content

Table 4 Comparison of AUC value for the empirical Bayes and other classifiers

From: Large-scale risk prediction applied to Genetic Analysis Workshop 17 mini-exome sequence data

Item

Model

Statistics

Empirical Bayes model

Random forest classifier

Neural network 1

Neural network 2

AUC value

Gene + environment

Mean

0.76

0.67

0.68

0.70

  

SE

0.0102

–

–

–

  1. AUC value indicates the area under the ROC curve when minimizing the cross-validation error. Neural network 1 used selected features from the logistic regression; neural network 2 used selected features from the empirical Bayes method. SE is the standard error of the AUC value.
Back to article page

BMC Proceedings

ISSN: 1753-6561

Contact us