Journal of Cloud Computing

Advances, Systems and Applications

Table 6 Model parameter summary information

From: MBi-GRUMCONV: A novel Multi Bi-GRU and Multi CNN-Based deep learning model for social media sentiment analysis

Hyperparameter	Values-Settings
Train size	70%, 80%
Test size	30%, 20%
Validation split	10%
Embedding Dimension	Word2Vec 100, 200, 300
Filter	128
Number of Units	128
Learning Rate	0.001
Optimizer	Adam
Batch Size	64
Number of Epochs	20
Recurrent Dropout	0.3
Max Length	200
Output Size	1
Metrics	Accuracy
Kernel Regularization	L2
Activation Function	ReLu
Loss Function	Binary Cross Entropy
Output Function	Sigmoid

Back to article page