Table 5. The hyperparameter values employed of our findings.

Hyperparameter	Value	Description
num_words	10,000	Maximum number of words to keep based on word frequency.
oov_token	<OOV>	Token to represent out-of-vocabulary words.
maxlen	100	Maximum length of sequences (padded/truncated).
embedding_dim	100	Dimensionality of the word embeddings.
input_dim	num_words	Size of the vocabulary.
output_dim	100	Dimensionality of the output space.
trainable	False	Whether the embedding layer is trainable.
filters	128	Number of filters in the convolutional layer.
kernel_size	5	Size of the convolutional kernel.
pool_size	4	Size of the max pooling window.
units	64	Number of units in the LSTM and dense layers.
dropout_rate	0.5	Fraction of input units to drop for dropout.
lr	0.001	Learning rate for the Adam optimizer.
batch_size	32	Number of samples per gradient update during training.
epochs	10	Number of epochs for training.

Notes.

The best performing results are shown in bold.