Research Article

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

Table 10

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

ModelEvaluation metricCross-validation performanceTesting performance
NormalGrayholeBlackholeSchedulingFloodingNormalGrayholeBlackholeSchedulingFlooding

DTTN99.9799.9699.9297.7499.8999.9699.9899.998.2399.89
FP0.030.040.082.260.110.040.020.11.770.11
FN0.944.731.80.217.11.183.111.440.236.07
Accuracy99.9599.9199.8699.699.7799.9399.9599.8599.6399.78
Precision98.9395.2498.1299.7794.1198.5697.3297.699.8293.74
Recall99.0695.2798.299.7992.998.8296.8998.5699.7793.93
F-score98.9995.2498.1699.7893.598.6997.198.0899.7993.83

KNNTN99.7199.8599.5684.2599.8999.7299.999.5784.1499.89
FP0.290.150.4415.750.110.280.10.4315.860.11
FN10.3525.3620.580.5335.259.8222.6721.010.4936.23
Accuracy99.4499.6398.7898.0699.2799.4699.7298.7598.0999.26
Precision89.4681.5788.0298.4191.6889.9585.7188.3598.491.57
Recall89.6574.6479.4299.4764.7590.1877.3378.9999.5163.77
F-score89.5577.9483.4898.9475.990.0781.3183.4198.9675.18

RFTN99.9699.9499.9598.2699.9999.9699.9699.9398.42100
FP0.040.060.051.740.010.040.040.071.580
FN0.461.361.420.127.230.520.891.530.126.07
Accuracy99.9599.9399.8999.7399.8699.9499.9599.8799.7499.89
Precision98.669498.6799.8299.5298.4595.398.399.8499.78
Recall99.5498.6498.5899.8892.7799.4899.1198.4799.8893.93
F-score99.0996.2698.6399.8596.0298.9697.1798.3999.8696.77

NBTN99.9793.0696.498.8190.5699.9793.0596.3299.0890.26
FP0.036.943.61.199.440.036.953.680.929.74
FN65.6918.5753.2116.9918.3165.3120.0052.2417.4215.99
Accuracy98.2192.9694.4784.4790.498.2092.9594.4084.1090.15
Precision97.149.4834.5399.8513.597.078.5134.9199.8913.37
Recall34.3181.4346.7983.0181.6934.6980.0047.7682.5884.01
F-score50.6716.9839.7290.6523.1751.1115.3840.3390.4123.07