Fig. 3From: Classifying breast cancer subtypes on multi-omics data via sparse canonical correlation analysis and deep learningNormalized confusion matrices of all competing methods on the breast cancer multi-omics dataset. In the confusion matrix of each method, the label of each row corresponds to the true label of breast cancer subtype and the label of each column represents the predicted label of breast cancer subtype. The diagonal entity in the matrix indicates the proportion of correctly predicted classes. The off-diagonal entity in the matrix indicates the proportion of misclassification. To account for imbalanced sample sizes of different breast cancer subtype dataset, the confusion matrices are normalized within the range of 0 to 1Back to article page