Table 6 Statistical Results of DP-RBFN vs. DP-MLP (10 Trials).

From: A service-oriented microservice framework for differential privacy-based protection in industrial IoT smart applications

\(\epsilon\)

DP-RBFN

DP-MLP

Accuracy

Std. dev

95% CI

Accuracy

Std. dev

95% CI

0.1

87.35

1.91

[86.10, 88.60]

85.21

2.34

[83.68, 86.74]

0.3

90.12

1.43

[89.17, 91.07]

88.06

2.01

[86.67, 89.45]

0.5

93.21

1.12

[92.47, 93.95]

91.87

1.34

[90.98, 92.76]

1.0

95.26

0.89

[94.63, 95.89]

94.17

1.21

[93.34, 95.00]

Back to article page