Skip to main content
. 2020 Aug 3;8:e9625. doi: 10.7717/peerj.9625

Table 2. The performance of the classification tasks by SNV-SVM, Random Forest, Naive Bayes and Blastn.

Accuracy Precision Recall F-measure AUC Time (s)
Genus classification SNV-SVM 0.990 0.987 0.992 0.990 0.998 0.553
Random Forest 0.980 0.974 0.988 0.981 0.997 3.158
Naive Bayes 0.867 0.871 0.896 0.884 0.923 1.717
Blastn 0.990 0.990 0.990 0.990 0.995 5.718
Genome component classification SNV-SVM 0.995 0.995 0.995 0.995 0.998 1.148
Random Forest 0.995 0.996 0.996 0.996 0.996 2.765
Naive Bayes 0.982 0.976 0.988 0.981 0.995 2.050
Blastn 1.000 1.000 1.000 1.000 1.000 3.745
Mono-or bipartite
DNA-A classification
SNV-SVM 0.987 0.985 0.991 0.988 0.998 0.293
Random Forest 0.952 0.947 0.953 0.952 0.980 1.862
Naive Bayes 0.977 0.967 0.987 0.977 0.973 0.219
Blastn 0.818 0.750 0.844 0.794 0.834 1.050