Algorithm 1. Proposed Model

Feature Optimization and CNN–GRU Input:       Data instances Output:       Confusion Matrix       (Accuracy, precision, recall, FPR, TPR) Dataset Optimization       Remove the redundant instances   Feature Selection Using Pearson’s Correlation equation, compute the correlation of the attribute set Set $C_f$.           if corr_value > 0.8                 add attribute to $C_f$ else                 increment in an attribute set $C$ return $C_f$ Classification       Create training and testing sets from the dataset.       Training set: 67%       Testing set: 33%       add model             three Convolution layers (activation = ‘relu’)             two GRU layers (activation = ‘relu’)       model compilation             loss function: ‘categorical_crossentropy’             optimizer=‘adagrad’       training CNN–GRU technique with training instances       employing techniques to test instances return Confusion Matrix $C_{m\ast m}$