Research Article
Driving Anger States Detection Based on Incremental Association Markov Blanket and Least Square Support Vector Machine
Table 5
The average classification performances of the six models for four driving anger states.
| ā | TPR | FPR | PPA | F1 | Acc | AUC |
| LSSVM | 0.8369 | 0.0749 | 0.8101 | 0.8215 | 0.8220 | 0.8764 | C4.5 | 0.7937 | 0.0490 | 0.7760 | 0.7828 | 0.8017 | 0.8623 | NBC | 0.7768 | 0.0627 | 0.7591 | 0.7653 | 0.7905 | 0.8439 | KNN | 0.7546 | 0.0698 | 0.7480 | 0.7487 | 0.7436 | 0.8275 | SVM | 0.7612 | 0.0582 | 0.7552 | 0.7562 | 0.7786 | 0.8376 | BPNN | 0.7408 | 0.0937 | 0.7458 | 0.7414 | 0.7384 | 0.8148 |
|
|