Table 5 Comparison of different classifiers when 5% of the class samples are used as gallery and remaining 95% of sample are used as test on 38 snake feature database
From: Discriminative histogram taxonomy features for snake species identification
Method | % Correct | F-score | AUC | Precision (%) | Recall (%) |
|---|---|---|---|---|---|
Bayes net [37] | 78.81 ± 2.27 | 0.98 ± 0.01 | 0.89 ± 0.03 | 83 ± 0.07 | 97 ± 0.03 |
Naïve Bayes [27] | 77.69 ± 2.11 | 0.97 ± 0.01 | 0.89 ± 0.02 | 90 ± 0.05 | 89 ± 0.03 |
Multilayer perceptron [26] | 86.85 ± 2.59 | 0.98 ± 0.01 | 0.92 ± 0.02 | 90 ± 0.04 | 94 ± 0.04 |
Ada BoostM1 [28] | 57.39 ± 1.44 | 0.80 ± 0.03 | 0.75 ± 0.02 | 62 ± 0.04 | 95 ± 0.05 |
Multi BoostAB [29] | 57.39 ± 1.44 | 0.80 ± 0.03 | 0.75 ± 0.02 | 62 ± 0.04 | 95 ± 0.05 |
RBF network [30] | 85.00 ± 3.05 | 0.95 ± 0.02 | 0.91 ± 0.03 | 92 ± 0.05 | 89 ± 0.04 |
IB1 [31] | 85.82 ± 2.45 | 0.93 ± 0.02 | 0.91 ± 0.02 | 88 ± 0.04 | 93 ± 0.04 |
IBk [31] | 86.38 ± 2.47 | 0.94 ± 0.01 | 0.91 ± 0.02 | 88 ± 0.04 | 94 ± 0.03 |
LWL [32] | 68.37 ± 6.26 | 0.97 ± 0.01 | 0.82 ± 0.05 | 72 ± 0.07 | 96 ± 0.03 |
NB Tree [36] | 83.79 ± 2.87 | 0.95 ± 0.02 | 0.91 ± 0.03 | 88 ± 0.05 | 95 ± 0.04 |
J48 [33] | 78.92 ± 4.37 | 0.98 ± 0.01 | 0.87 ± 0.04 | 80 ± 0.06 | 97 ± 0.03 |
Random sub space [34] | 80.50 ± 3.29 | 0.97 ± 0.02 | 0.90 ± 0.03 | 85 ± 0.06 | 96 ± 0.04 |
Bagging [35] | 80.91 ± 4.48 | 0.94 ± 0.03 | 0.88 ± 0.04 | 83 ± 0.06 | 93 ± 0.04 |