A Survey on COVID-19 Future Forecasting using Machine Learning Models

Abstract

Abstract: Predictive methods for Learning Machines (MLs) have proved their worth in expecting long-term performance results to improve decision-making over the course of future action. ML models have long been used in many application domains that need to be identified and prioritized against negative threats. Many prediction methods are widely used to manage prediction problems. This study demonstrates the ability of ML models to predict the future of affected patients by COVID-19 and they are currently considered to be a threat to humanity. Predictions are made by each Machine Learning model to determine the number of confirmed cases in upcoming days. In this paper, the review is undertaken for a few methodologies that are applied to the solution of COVID19 problem. The ML model architecture is proposed with a day wise and country wise analysis for confirmed cases of COVID19 problem.

Index Terms: COVID-19, Machine Learning methods, Time series methods future forecasting, adjusted R2 score, supervised machine learning