Table 2. The mean and standard deviation TPR for the anomaly detection algorithms.

The results represent experiments on four datasets based on 20 simulated experiments. The proposed method (NewAlgo) produces the best overall results.

	DS1	DS2	DS3	DS4	Overall
Naive	1.000±0.00	0.780±0.112	0.490±0.158	1.000±0.000	0.818
RobustCov	0.975±0.043	0.420±0.194	0.020±0.068	0.450±0.470	0.466
IsoForest	0.24±0.132	0.000±0.000	0.000±0.000	0.015±0.036	0.065
LOF	0.040±0.058	0.045±0.074	0.055±0.086	0.055±0.092	0.049
NewAlgo	0.915±0.111	0.760±0.097	0.840±0.107	0.980±0.051	0.874