Regression Analysis And Future Forecasting Of COVID-19 Using Machine Learnings Algorithm

• Published in: 2021 11th International Conference on Cloud Computing, Data Science & Engineering (Confluence) pp. 1014 - 1018
• Main Authors: Gera, Saksham, Mridul, Mr, Joshi, Kireet
• Format: Conference Proceeding
• Language: English
• Published: IEEE 28.01.2021
• Subjects: Corona virus; COVID-19; Exponential smoothing; Forecasting; Pandemics; Predictive models; Random forest; Random forests; Regression; Regression analysis; Support vector machines
• DOI: 10.1109/Confluence51648.2021.9377065

The COVID -19 (Novel Corona Virus) Pandemic has strike the world and cause a great destruction in life. It is considered as one of the disastrous Pandemic in the history of the world. This paper aims to gives insight of how different models of ML is contrivance in current situation. In addition to the regression analysis performed on Indian data, the study examines contemporaneous pattern or trend in COVID-19 transmission in India. Also, forecasting system based on Machine Learning has shown its importance for enhancement of the managerial ability on ensuing course of action. This research demonstrates the ability of different models of Machine Learning to prognosticate the number of coming patients affected by nCov, a ultimatum to mankind. In this study 5 models: LR, SVM, Random Forest, KNN and ES have been used. Two types of prediction are made by each model: 1) number of newly positive confirm eases 2) number of deaths. This study proves that among all models ES perform best, followed by Random Forest and KNN which perform better than SVM that perform poorly in all prediction areas.