An Optimized Hyperparameter of Convolutional Neural Network Algorithm for Bug Severity Prediction in Alzheimer’s-Based IoT System

Algorithm 1

Pseudocode of the proposed approach hybrid CNN-HHO.

(1)	function CNN–HHO (x, y, k-value, N, T)\\The function takes size of population N and, the number of iteration T, train-test split dataset with target value
(2)	32 and 64\\The batch size of DL model
(3)	Adam, Adadelta, Adagrad, Adamax, NAdam, SGD, RMSprop\\DL Optimizers
(4)	0.01 to 0. \\The learning rate of DL model
(5)	softmax, softplus, softsign, relu, tanh, sigmoid, hard_sigmoid, linear\\Activation function
(6)	uniform, lecun_uniform, normal, zero, glorot_normal, he_normal, glorot_uniform, he_uniform∖∖Neural network kernel initializers
(7)	Accuracy, Precision, Recall, F1-Measure\\Evaluation metrics
(8)	CNN model\\DL Model
(9)	1\\initial iteration
(10)	Pop  initiate pop (N)\\initiate the population with pop size
(11)	while do
(12)	train-test split dataset\\dataset splitting
(13)	popScores  Evaluate the fitness value (M, , )
(14)	newPop  updatePop (popScores, t, T)\\update population
(15)	pop  newPop\\set new population
(16)	t  t + 1∖∖ iteration update
(17)	best Solution  Extract Top (Pop)\\Extract best solution
(18)	Result  Best hyper parameters and Best solution \\Result with best solutions and hyper parameters