Table 15.
Model | Hyperparameters |
---|---|
LSTM | Embedding (4000, 100, input_length = …) Dropout (0.5) LSTM (128) Dense (5, activation = ‘softmax’) |
CNN | Embedding (4000, 100, input_length = …) Conv1D (128, 5, activation = ‘relu’) MaxPooling1D (pool_size = 5) Activation (‘relu’) Dropout (rate = 0.5) Flatten() Dense (5, activation = ‘softmax’) |
CNN-LSTM | Embedding (4000, 100, input_length = …) Conv1D (128, 5, activation = ‘relu’) MaxPooling1D (pool_size = 5) LSTM (100) Dense (5, activation = ‘softmax’) |
loss = ‘categorical_crossentropy’, optimizer = ‘adam’, epochs = 100, batch_size = 16 |