Autonomous Classification and Decision-Making Support of Citizen E-Petitions Based on Bi-LSTM-CNN

Table 6

Key parameters of the classification model to be optimized.

Parameter	Explanation	Scope
Batch size	Number of samples per gradient update
Learning rate	Hyperparameters in updating weights during gradient descent
Embedding size	The dimensionality of the word vector
Hidden size	Number of neurons in the hidden layer
Kernel size	Length of convolution window in Bi-LSTM-CNN layer
Kernel size	Length of convolution window in CNN layer
Filter size	Number filters in Bi-LSTM-CNN layer
Filter size	Number filters in CNN layer
Dropout	A simple way to prevent neural networks from overfitting
Activation function	A function for introducing nonlinear factors into a neural network	Relu or tanh