Hybridization of Neural Networks and Sine Cosine Algorithm for an Optimal Neural Network Architecture Applied to Prevent Heart Attacks

Abstract

Artificial intelligence and deep learning provide very good results, if it is well adjusted. In this work, we will proceed to perform the deep learning results through the optimization of the neural network architecture. For this purpose, especially for supervised algorithms, Hybridization of neural networks and the sinus-cosine algorithm will perform classification problems. The role of this method is to escape the groping method in choosing the optimal architecture of neural networks. The goal of our method is to build an optimal neural network architecture, without falling into an over fitting problem. To demonstrate the effectiveness of our work, an application part with experimental results is included: with an application in medicine especially heart attack. The goal of our work is to develop an efficient hybrid classifier, using machine learning and sine cosine algorithm to detect heart attack and minimize the number of heart attack not predicted. Through this hybridization technique we should have a low error with satisfying classification results. This method of hybridization can be applied for different issues.