Skip to main content
2023 Jun 3:1–31. Online ahead of print. doi: 10.1007/s42600-023-00286-8

Table 4.

Sample mean and standard error for accuracy, sensitivity, specificity, and area under ROC curve (AUC) for all classifiers, considering the 8-feature PSO dimension-reduced dataset.

Source: Authors

Classifier SARS-CoV-2 PSO
Accuracy Kappa Sensitivity Specificity AUC F1 score
Mean Std dev Mean Std dev Mean Std dev Mean Std dev Mean Std dev Mean Std dev
Naïve Bayes 99.9689 0.1369 0.9963 0.0149 0.9997 0.0014 1.0000 0.0000 1.0000 0.0001 0.9998 0.0007
Bayes Network 100.0000 0.0000 1.0000 0.0000 1.0000 0.0000 1.0000 0.0000 1.0000 0.0000 1.0000 0.0000
J48 tree 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
Random tree 99.9539 0.0849 0.9942 0.0107 0.9998 0.0006 0.9939 0.0151 0.9969 0.0076 0.9998 0.0004
Random forest (10 trees) 99.9812 0.0517 0.9977 0.0064 0.9998 0.0005 0.9992 0.0054 1.0000 0.0001 0.9999 0.0003
Random forest (20 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (30 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (40 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (50 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (60 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (70 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (80 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (90 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
Random forest (100 trees) 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 1.0000 0.0000 0.9999 0.0003
MLP (20 neurons) 99.9496 0.0845 0.9937 0.0106 0.9997 0.0007 0.9955 0.0132 0.9999 0.0005 0.9997 0.0004
MLP (50 neurons) 99.9458 0.0888 0.9932 0.0111 0.9997 0.0007 0.9946 0.0145 0.9999 0.0005 0.9997 0.0005
MLP (100 neurons) 99.9458 0.0888 0.9932 0.0112 0.9997 0.0007 0.9943 0.0154 0.9999 0.0005 0.9997 0.0005
SVM polynomial E1; C = 0.01 96.3626 0.5919 0.1952 0.1561 1.0000 0.0001 0.1240 0.1440 0.5620 0.0720 0.9814 0.0030
SVM polynomial E2; C = 0.01 95.9281 0.4686 0.0245 0.1154 1.0000 0.0002 0.0195 0.1152 0.5098 0.0575 0.9792 0.0024
SVM polynomial E3; C = 0.01 95.9078 0.4678 0.0160 0.1137 1.0000 0.0002 0.0149 0.1150 0.5074 0.0574 0.9791 0.0024
SVM polynomial E4; C = 0.01 95.9024 0.4675 0.0136 0.1133 1.0000 0.0002 0.0136 0.1149 0.5068 0.0574 0.9791 0.0024
SVM polynomial E5; C = 0.01 95.8890 0.4708 0.0131 0.1133 0.9998 0.0005 0.0135 0.1149 0.5066 0.0574 0.9790 0.0024
SVM RBF G = 0.01; C = 0.01 95.8488 0.0629 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.5000 0.0000 0.9788 0.0003
SVM RBF G = 0.25; C = 0.01 95.9019 0.4631 0.0130 0.1125 1.0000 0.0000 0.0128 0.1105 0.5064 0.0552 0.9791 0.0024
SVM RBF G = 0.5; C = 0.01 95.9067 0.4728 0.0148 0.1146 1.0000 0.0001 0.0142 0.1151 0.5071 0.0575 0.9791 0.0024
SVM polynomial E1; C = 0.1 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
SVM polynomial E2; C = 0.1 99.9673 0.0687 0.9960 0.0084 0.9997 0.0007 1.0000 0.0000 0.9998 0.0004 0.9998 0.0004
SVM polynomial E3; C = 0.1 99.9673 0.0687 0.9960 0.0084 0.9997 0.0007 1.0000 0.0000 0.9998 0.0004 0.9998 0.0004
SVM polynomial E4; C = 0.1 96.1480 0.4815 0.1227 0.1305 0.9998 0.0005 0.0763 0.1190 0.5381 0.0594 0.9803 0.0025
SVM polynomial E5; C = 0.1 96.0097 0.4765 0.0659 0.1237 0.9998 0.0005 0.0430 0.1171 0.5214 0.0585 0.9796 0.0024
SVM RBF G = 0.01; C = 0.1 95.8488 0.0629 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.5000 0.0000 0.9788 0.0003
SVM RBF G = 0.25; C = 0.1 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
SVM RBF G = 0.5; C = 0.1 99.9689 0.0663 0.9961 0.0083 0.9998 0.0005 0.9964 0.0113 0.9981 0.0057 0.9998 0.0003
SVM polynomial E1; C = 1 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
SVM polynomial E2; C = 1 99.9673 0.0687 0.9960 0.0084 0.9997 0.0007 1.0000 0.0000 0.9998 0.0004 0.9998 0.0004
SVM polynomial E3; C = 1 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM polynomial E4; C = 1 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM polynomial E5; C = 1 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM RBF G = 0.01; C = 1 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
SVM RBF G = 0.25; C = 1 99.9684 0.0667 0.9960 0.0084 0.9998 0.0005 0.9963 0.0115 0.9980 0.0057 0.9998 0.0003
SVM RBF G = 0.5; C = 1 99.9678 0.0671 0.9960 0.0084 0.9998 0.0005 0.9961 0.0117 0.9980 0.0058 0.9998 0.0003
SVM polynomial E1; C = 10 99.9689 0.0676 0.9962 0.0083 0.9997 0.0007 1.0000 0.0000 0.9998 0.0004 0.9998 0.0004
SVM polynomial E2; C = 10 99.9657 0.0711 0.9958 0.0087 0.9996 0.0007 1.0000 0.0000 0.9998 0.0004 0.9998 0.0004
SVM polynomial E3; C = 10 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM polynomial E4; C = 10 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM polynomial E5; C = 10 99.9517 0.0816 0.9941 0.0100 0.9995 0.0009 1.0000 0.0000 0.9997 0.0004 0.9997 0.0004
SVM RBF G = 0.01; C = 10 99.9839 0.0483 0.9980 0.0059 0.9998 0.0005 1.0000 0.0000 0.9999 0.0003 0.9999 0.0003
SVM RBF G = 0.25; C = 10 99.9807 0.0540 0.9976 0.0067 0.9998 0.0005 0.9992 0.0054 0.9995 0.0027 0.9999 0.0003
SVM RBF G = 0.5; C = 10 99.9684 0.0667 0.9960 0.0084 0.9998 0.0005 0.9963 0.0115 0.9980 0.0057 0.9998 0.0003