Securing the IoT System of Smart City against Cyber Threats Using Deep Learning

Table 2

Parameters details.

Name	Parameter
Test size	0.2 (for all classifier)
Number of trees	1000 (for random forest) and 500 (for majority voting)
Loss	Categorical cross-entropy (for ANN/CNN)
Optimizer	Adam (for ANN/CNN)
Dropout	0.5 (for ANN/CNN)
Batch size	20 (for ANN) and 128 (for CNN)
Epochs	15 (for ANN) and 20 (for CNN)