Research Article
A Hybrid Swarm Intelligence Algorithm for Intrusion Detection Using Significant Features
Table 9
Accuracy rates of classifiers using SFSM feature selection method and Friedman ranks.
| | Dataset | NB | J4.8 | RBF | SVM | ABC | MABC-EPSO |
| | DoS + 10% normal | 82.57 () | 87.11 () | 87.96 () | 84.7 () | 90.82 () | 99.50 () | | Probe + 10% normal | 82.68 () | 82.6 () | 83.72 () | 85.67 () | 96.58 () | 99.27 () | | R2L + 10% normal | 86.15 () | 82.55 () | 85.16 () | 90.61 () | 92.72 () | 99.24 () | | U2R + 10% normal | 84.06 () | 87.16 () | 85.54 () | 85.97 () | 97.31 () | 99.8 () |
| | Average rank | 5.25 | 4.75 | 4.25 | 3.75 | 2 | 1 |
|
|
