Abstract and Applied Analysis / 2014 / Article / Tab 5 / Research Article
Nonlinear Methodologies for Identifying Seismic Event and Nuclear Explosion Using Random Forest, Support Vector Machine, and Naive Bayes Classification Table 5 Discriminant results of test samples.
Number Event Measured results NBC SVM result RF Prob. (N) Prob. (NE) Result Prob. (N) Prob. (NE) Result 1 NE20 2 0.1646 0.8354 2 2 0.2354 0.7646 2 2 NE21 2 0.0276 0.9724 2 2 0.018 0.982 2 3 NE22 2 0.0041 0.9959 2 2 0.1572 0.8428 2 4 NE23 2 0.0144 0.9856 2 2 0.0178 0.9822 2 5 NE24 2 0.2324 0.7676 2 2 0.3342 0.6658 2 6 NE25 2 0.4416 0.5584 2 2 0.0744 0.9256 2 7 NE26 2 0.2447 0.7553 2
0.1576 0.8424 2 8 NE27 2 0.7336 0.2664
0.4752 0.5248 2 9 E15 1 0.998 0.002 1 1 0.8934 0.1066 1 10 E16 1 0.9991 0.0009 1 1 0.8214 0.1786 1 11 E17 1 0.8956 0.1044 1 1 0.7562 0.2438 1 12 E18 1 0.9807 0.0193 1 1 0.4076 0.5924
13 E19 1 0.9277 0.0723 1 1 0.8904 0.1096 1 14 E20 1 0.9725 0.0275 1 1 0.9216 0.0784 1
Note. result with ā
ā is incorrect.