NNA and Activation Equation-Based Prediction of New COVID-19 Infections

Abstract

At 2019, China had a large number of severe cases of pneumonia, particularly in Wuhan. A SARS virus was detected after a thorough realization of sample from the sick people. Due to the form of the virus, which resembled a crown, it was given the name CORONA; the abbreviation COVID-19 stands for 2019 CORONA VIRUS. The World Health Organization WHO classified it as COVID-19, a pandemic, on March, 2020. In this study, artificial neural networks—which function similarly to the network of human neurons—are built to imitate how the human brain functions. Due to this, neural networks were used to connect the diagnosis to the symptoms, where the platform and knowledge-based system were found to be compatible, the symptoms that depend on the diagnosed disease were represented as numerical data, and after the network had been trained, the system was found to be appropriate for the accurate diagnosis of the disease. Our current study includes two primary phases: the training phase of neurons, which includes inputting the training data and generating random weights whose value is less than 1 for each of these inputs, and applying the neural network algorithm to them. The testing phase, where the two inputs were entered without the results to assess how well the proposed system works. Three statistical calculations R, RMSE, MAPE were made in order to evaluate the performance of the existing system and its findings.