Table 6 Model training performance for UNSW-NB15 dataset.

From: Securing IoT networks: a machine learning approach for detecting unusual traffic patterns

Metric

Value before

tuning

Value after

tuning

Loss function

0.35

0.15

Accuracy

94.0%

98.7%

Precision

93.5%

98.3%

Recall

92.8%

99.0%

F1 Score

93.1%

98.6%