Research Article

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

Table 11

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

ModelEvaluation metricCross-validation performanceTesting performance
NormalGrayholeBlackholeSchedulingFloodingNormalGrayholeBlackholeSchedulingFlooding

DTTN99.9599.9599.997.7299.9199.9699.9699.9197.4199.91
FP0.050.050.12.280.090.040.040.092.590.09
FN1.325.182.320.226.871.196.642.550.217.17
Accuracy99.9299.999.8299.5999.7999.9299.9099.8199.5799.78
Precision98.2994.1697.6599.7794.7698.3794.9697.7599.7494.77
Recall98.6894.8297.6899.7893.1398.8193.3697.4599.7992.83
F-score98.4894.4897.6699.7793.9398.5994.1597.6099.7693.79

KNNTN99.5999.8199.3880.1699.9899.6399.8199.3780.6099.98
FP0.410.190.6219.840.020.370.190.6319.400.02
FN16.1332.1127.180.640.5516.2428.7427.270.5938.31
Accuracy99.1799.5398.3497.6299.2699.2199.5598.3397.6799.30
Precision84.8476.4682.5598.0197.786.2676.5282.3298.0598.27
Recall83.8767.8972.8299.459.4583.7671.2672.7399.4161.69
F-score84.3471.8877.3698.773.9184.9973.8077.2398.7275.80

RFTN99.9599.9399.9398.2899.9699.9599.9499.9598.2899.96
FP0.050.070.071.720.040.050.060.051.720.04
FN0.910.961.890.177.040.600.722.030.156.99
Accuracy99.9399.9299.8699.6999.8499.9399.9499.8799.7199.84
Precision98.1692.8198.2499.8297.8498.1193.9498.6899.8297.66
Recall99.0999.0498.1199.8392.9699.4099.2897.9799.8593.01
F-score98.6295.8298.1799.8395.3398.7596.5498.3299.8495.28

NBTN99.4891.6296.0456.0199.9599.5091.4995.9557.4299.96
FP0.528.383.9643.990.050.508.514.0542.580.04
FN63.8625.0970.412.4840.9361.8226.4570.8712.5840.60
Accuracy97.7891.4793.4584.6199.2397.8591.3393.3484.6599.24
Precision65.477.3923.2695.1495.8367.687.1622.5595.2896.20
Recall36.1474.9129.687.5259.0738.1873.5529.1387.4259.40
F-score46.5413.4426.0491.1773.0848.8213.0525.4291.1873.45