Research Article
Detecting Cyber-Attacks on Wireless Mobile Networks Using Multicriterion Fuzzy Classifier with Genetic Attribute Selection
Table 5
The WEP/WPA per-class performance of the proposed method with and without attribute selection (approximated to three decimal digits).
| Normal/attack | Count | With attribute selection | Without attribute selection | Precision | Recall | | AUC | Precision | Recall | | AUC |
| Normal | 15000 | 0.822 | 1 | 0.902 | 1 | 0.817 | 0.996 | 0.898 | 0.993 | ChopChop | 2300 | 1 | 0.326 | 0.491 | 0.627 | 0.855 | 0.130 | 0.226 | 0.521 | Deauthentication | 2300 | 0.945 | 0.971 | 0.958 | 0.984 | 0.938 | 0.981 | 0.959 | 0.989 | Duration | 2300 | 0.970 | 0.997 | 0.983 | 0.999 | 0.966 | 0.986 | 0.976 | 0.992 | Fragmentation | 2300 | 0.994 | 0.21 | 0.347 | 0.563 | 0.968 | 0.338 | 0.50 | 0.632 |
|
|