Table 1.

Hyperparameters of gated recurrent unit, bidirectional long short-term memory, and convolutional neural network.

Hyperparametera	GRUb	BiLSTMc	CNNd
Filter	—e	—	32
Kernel	—	—	4, 6, 8
Activation function (last layer)	ReLUf (Softmax)	ReLU (Softmax)	ReLU (Sigmoid)
Dropout	0.5	0.5	0.5
Epoch	50	50	50
Batch size	128	128	128
Loss function	categorical_crossentropy	categorical_crossentropy	binary_crossentropy
Optimizer/learning rate	Adam/0.0001	Adam/0.0001	Adam/0.0001
Regularization	Batch normalization	Batch normalization	Batch normalization

^aBiLSTM: bidirectional long short-term memory.

^bGRU: gated recurrent unit.

^cCNN: convolutional neural network.