Research Article

Trust in Intrusion Detection Systems: An Investigation of Performance Analysis for Machine Learning and Deep Learning Models

Table 15

The performance of machine learning for KDD dataset using selected features.

ModelsEvaluation metricCross-validation performanceTesting performance
DosNormalProbeR2LU2RDOSNormalProbeR2LU2R

DTTNR98.7599.9599.9810010098.9199.9599.9799.99100.00
FPR1.250.050.02001.090.050.030.010.00
FNR0.060.1725.372.450.830.060.1822.762.0575.00
Accuracy99.7299.9399.7699.9999.9999.7499.9399.7899.9999.99
Precision99.7199.7696.7598.4375.9399.7499.7795.6097.9542.86
Recall99.9499.8674.6397.649.1799.9499.8277.2497.9525.00
F-score99.8399.884.259857.7899.8499.8085.4597.9531.58

KNNTNR98.5599.8799.9799.9910098.7299.8699.9899.99100.00
FPR1.450.130.030.0101.280.140.020.010.00
FNR0.070.3827.6516.111000.070.3325.0719.94100.00
Accuracy99.6799.8399.7399.9599.9999.7099.8399.7699.9499.99
Precision99.6699.4195.8996.80.099.7099.3796.4594.790.0
Recall99.9399.6272.3583.89099.9399.6774.9380.060.00
F-score99.899.5182.4689.810.099.8199.5284.3486.800.0

RFTNR98.7699.9499.9810010098.9199.9599.97100.00100.00
FPR1.240.060.02001.090.050.030.000.00
FNR0.060.1425.293.08400.060.1322.313.8158.33
Accuracy99.7299.9399.7699.9910099.7499.9499.7899.9999.99
Precision99.7199.7696.6699.7487.0499.7499.7796.1599.3971.43
Recall99.9499.8674.7196.926099.9499.8777.6996.1941.67
F-score99.8399.884.2698.371.8599.8499.8285.9497.7652.63

NBTNR21.699.7597.6599.4979.4721.2799.7299.7299.5277.36
FPR78.40.252.350.5120.5378.730.280.280.4822.64
FNR24.797.3592.9798.6420.8324.9096.72100.00100.008.33
Accuracy65.1382.4696.8799.2479.4764.8782.4898.8799.2777.36
Precision80.4466.470.290.480.0980.2771.770.000.000.04
Recall75.32.657.031.3679.1775.103.280.000.0091.67
F-score77.785.085.613.560.1977.606.27000.07