---

Corona Virus Disease 2019 (COVID-19): Prediction Strategy Using Sequential Deep Learning Model

Corresponding Author : A. S. Surja (amitshaha719@gmail.com)

Authors : A. S. Surja (amitshaha719@gmail.com), M. O. Faruk (omorsec@gmail.com), M. S. Iqbal (shahid_05@yahoo.com)

Keywords : ANN, Deep Learning, Sequential Model, Backpropagation, Covid-19

Abstract :

Nowadays, while the globe is facing extreme difficulty with COVID-19, Artificial Intelligence can help to cope with this epidemic in an innumerable number of ways. Motivated by this, in this article, a robust prediction model has been proposed for pandemic COVID-19 using Sequential Deep Learning (SDL), called COVID-SDL to predict the total positive cases. The proposed model utilizes the correlation information among the features that have strengthened the prediction capability. In addition, to assist the prediction ability of COVID-SDL, ReLu (Rectified Linear Unit) activation function has been used which enhanced the robustness of the model. With a view to making the predictions highly accurate, Adam optimizer has been adopted which works by reducing the cost function and making further updates of the weights. In order to evaluate the performance of COVID-SDL, data samples used in the model have been collected from Italy’s COVID-19 situation reports. Besides this, the dataset has gone through the processes of cleaning, filtering, formatting and visualization. Also, the exploratory survey shows that five most salient features (Home Confinement, Deaths, Recovered, Current Positive Cases and Tests Performed) results better which is obtained from Italy’s COVID-19 situation reports that are primarily composed of 16 features. Furthermore, COVID-SDL successfully obtained 0.00037316 MAE, 0.00000018 MSE, 0.00043476 RMSE and 0.99999 R2 Score with providing the best fit curve of predicted data which covers 99.999% of the actual data. Furthermore, to prove the robustness of the COVID-SDL, a comparative test among the Gradient Descent and Adam optimizers has also been performed.

Published on March 9th, 2023 in Volume 4 Issue 1, Computer Science, Electrical and Electronics