Research Article

DDoS Attack Detection and Classification Using Hybrid Model for Multicontroller SDN

Table 2

Comparison selective of four proposed methods.

Algorithm typeAttack typePerformance matrices
Accuracy (%)PrecisionRecallF1-score

RNN (with feature selection)BENIGN98.60.000.000.00
DrDoS_LDAP1.001.001.00
DrDoS_UDP0.991.001.00
SYN0.970.990.98
UDP-lag0.970.980.97
WebDDoS0.000.000.00

GRU (with feature selection)BENIGN96.40.000.000.00
DrDoS_LDAP1.001.001.00
DrDoS_UDP0.900.990.94
SYN1.000.970.99
UDP-lag0.980.910.94
WebDDoS0.000.000.00

MLP (with feature selection)BENIGN98.30.960.340.50
DrDoS_LDAP0.991.000.99
DrDoS_UDP0.980.980.98
SYN1.000.980.99
UDP-lag0.970.980.98
WebDDoS0.000.000.00

LSTM with feature selection (proposed solution)BENIGN99.420.780.630.70
DrDoS_LDAP1.001.001.00
DrDoS_UDP0.991.001.00
SYN1.000.991.00
UDP-lag1.000.990.99
WebDDoS0.000.000.00