Table 4 Comparison of different classifiers when 5% of the class samples are used as gallery and remaining 95% of sample are used as test on top 15 selected snake feature database
From: Discriminative histogram taxonomy features for snake species identification
Method | % Correct | F-score | AUC | Precision (%) | Recall (%) |
|---|---|---|---|---|---|
Bayes net [37] | 81.26 ± 4.00 | 0.98 ± 0.01 | 0.92 ± 0.03 | 88 ± 0.05 | 96 ± 0.04 |
Naïve Bayes [27] | 81.64 ± 3.05 | 0.98 ± 0.01 | 0.93 ± 0.03 | 91 ± 0.04 | 96 ± 0.04 |
Multilayer perceptron [26] | 86.64 ± 2.71 | 0.97 ± 0.01 | 0.92 ± 0.02 | 90 ± 0.04 | 95 ± 0.03 |
Ada BoostM1 [28] | 57.52 ± 1.27 | 0.80 ± 0.03 | 0.75 ± 0.02 | 63 ± 0.04 | 95 ± 0.04 |
Multi BoostAB [29] | 57.52 ± 1.27 | 0.80 ± 0.03 | 0.75 ± 0.02 | 63 ± 0.04 | 95 ± 0.04 |
RBF network [30] | 88.75 ± 2.69 | 0.97 ± 0.02 | 0.94 ± 0.02 | 93 ± 0.04 | 96 ± 0.03 |
IB1 [31] | 86.05 ± 3.35 | 0.93 ± 0.03 | 0.91 ± 0.04 | 89 ± 0.04 | 94 ± 0.07 |
IBk [31] | 87.50 ± 2.35 | 0.95 ± 0.01 | 0.92 ± 0.02 | 88 ± 0.04 | 96 ± 0.03 |
LWL [32] | 69.57 ± 4.17 | 0.97 ± 0.01 | 0.86 ± 0.04 | 77 ± 0.06 | 96 ± 0.03 |
J48 [33] | 84.71 ± 2.90 | 0.95 ± 0.02 | 0.91 ± 0.03 | 87 ± 0.06 | 96 ± 0.03 |
Random sub space [34] | 79.77 ± 4.01 | 0.98 ± 0.01 | 0.89 ± 0.03 | 82 ± 0.06 | 98 ± 0.02 |
Bagging [35] | 81.34 ± 3.93 | 0.97 ± 0.02 | 0.90 ± 0.03 | 85 ± 0.06 | 96 ± 0.03 |
NB Tree [36] | 82.10 ± 4.02 | 0.96 ± 0.02 | 0.91 ± 0.03 | 87 ± 0.05 | 96 ± 0.03 |