Skip to main content

Table 2 The performances of CapsNet-SSP and other methods on the independent test set

From: CapsNet-SSP: multilane capsule network for predicting human saliva-secretory proteins

Methods Accuracy Sensitivity Specificity Precision F-score MCC AUC
KNN 0.778 0.649 0.907 0.875 0.745 0.575 0.809
Decision Tree 0.772 0.692 0.851 0.823 0.752 0.550 0.740
Random Forest 0.781 0.804 0.758 0.769 0.786 0.563 0.836
AdaBoost 0.792 0.703 0.881 0.855 0.772 0.593 0.847
SVM 0.781 0.784 0.778 0.779 0.782 0.562 0.857
CapsNet-SSP 0.888 0.847 0.929 0.922 0.884 0.779 0.948
  1. The threshold is set where the MCC reaches the maximum value