COVID-19 ANALYSIS USING MACHINE LEARNING AND DEEP LEARNING ALGORITHMS

Abstract

COVID-19 or by another name known as coronavirus is a disease caused by Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2). At first, this disease originated from Wuhan, Hubei Province which then developed and spread suddenly throughout the world. COVID-19 has had a severe and widespread impact, especially in Indonesia. COVID-19 was first reported in Indonesia on 03 March 2020, then spread rapidly to 34 provinces on 09 April 2020. Artificial intelligence researchers focused their knowledge of expertise to develop mathematical models to analyze this epidemic situation using shared data nationwide. To contribute to the welfare of the living community, this research proposes to utilize machine learning and deep learning models to understand everyday exponential behavior along with predictions of the future reach of COVID-2019 in Indonesia. The amount of data used reached around 7098 data from March 1 to October 11, 2020, in 34 provinces of Indonesia. The use of this dataset aims to group each province in Indonesia into certain clusters so that it can identify areas with a large number of cases and a relatively small number of cases. Feature correlations are categorized based on the main island in Indonesia. The results of these two approaches are compared to determine the best-fitted model. In conclusion, machine learning prediction produces a better result than deep learning.