Covid-19 analysis by using machine and deep learning

Coronavirus is the pandemic in the whole world due to infection spread with community transfer. World Health Organization renames coronavirus to COVID-19, and the full name of coronavirus is severe acute respiratory syndrome coronavirus (SARS-CoV). COVID-19 belongs to the Betacoronavirus family and affects the respiratory system of humans. Machine learning is the part of artificial intelligence which used existing machine learning algorithms and dataset pattern to find an adequate solution for the problem. This chapter used machine learning algorithms supervised and unsupervised to analyze the spreading pattern such as confirmed case, recovered case, and death case of coronavirus worldwide. Analysis of the infection rate and mortality rate of coronavirus in the top 10 countries of the world. Analysis the effect of coronavirus in Asia and Europe visualized the result on the map. Choropleth is a thematic map which divides the world geographical area into different colors on the base of the data variable. This chapter used machine learning algorithms such as support vector machine, navies Bayes theorem, linear regression, decision tree repressors, random forest, and prophet algorithm for future prediction and to test the accuracy of prediction. The analysis of confirmed, recovered, and death cases is visualized by using Matplotlib in Python. Time series analysis of recovered, confirmed, and death cases of coronavirus predicted the future infection of the virus on the world. Analysis of the effect of Coronavirus, Country-wise, and State wise on top infected countries. Deep learning algorithm, long short-term memory (LSTM), is a recurrent neural network-type algorithm used to predict the future infection rate of coronavirus in the world. In this study, different machine learning algorithms were implemented and find the algorithm which gives the highest accuracy. Bar graph and pie are used to visualize the result of the experiment by using Python which helps to better understand the result. Analysis of the effect of coronavirus on Russia, Italy, the USA, the UK, Iran, Turkey, Germany, France, and Brazil finds the infection rate, mortality rate, and recovery rate of these countries. Analyze the precision, recall, and f1-score of these algorithms, and plot the graph. Find the recovered and death rates per one thousand patients of confirmed cases. © Springer International Publishing AG 2018.