Skip to main content

Table 3 Dataset D: classification performance of Ph-CNN compared to other classifiers on Healthy vs. UCr classification task

From: Phylogenetic convolutional neural networks in metagenomics

UCr Ph-CNN LSVM
p MCC min CI max CI MCC min CI max CI
60 0.861 0.855 0.867 0.811 0.807 0.815
119 0.893 0.888 0.899 0.866 0.862 0.870
178 0.906 0.900 0.911 0.892 0.888 0.895
237 0.920 0.916 0.924 0.917 0.914 0.920
  MLPNN RF
p MCC min CI max CI MCC min CI max CI
60 0.873 0.869 0.443 0.797 0.792 0.801
119 0.877 0.873 0.877 0.799 0.794 0.803
178 0.859 0.855 0.880 0.791 0.787 0.794
237 0.849 0.844 0.854 0.790 0.786 0.795
  1. The performance measure is MCC, with 95% studentized bootstrap confidence intervals (min CI, max CI). Models are computed for p={25%,50%,75% and 100%} of total number of features for each task. Comparing algorithms are linear Support Vector Machines (LSVM), Random Forest (RF) and MultiLayer Perceptron (MLPNN)