Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Win percentage: a novel measure for assessing the suitability of machine classifiers for biological problems

Figure 1

Classifier discrimination plots for breast cancer, pathological complete response. Panels A, B, C, D, E, and F correspond to the classifiers, NC, SDA, DLDA, UDA, LDA, and QDA, respectively. RD indicates residual invasive tumor, and pCR indicates pathological complete response. The black line indicates the average decision boundary across 100 folds of cross-validation. The cyan shading indicates the uncertainty in the labeling. White areas are labeled the same for every fold, whereas dark cyan represents the most uncertainty. The ellipses correspond to one standard deviation away from the mean for each class indicated with an 'x'.

Back to article page