Machine Learning based Trend analysis and Forecasting of COVID-19 using LSTM

Abstract

Humanity is in peril due to Covid's (the Coronavirus) activity in every country. Due to this illness' extreme infectiousness and infectivity, the resources of the world's largest countries are under stress. The ability of AI models to bet on the volume and variety of impending human agencies affected by Covid, which was eventually interpreted as a potential chance for humanity. Covid's subverting components were evaluated particularly using Three layer of deciding styles, least overall shrinkage, and decision manager (Rope) Backing vector Machine - deep learning. Least All of the frameworks propose three kinds of assumptions: the number of recently spoiled reviews, the number of passings, and the number of recoveries. Patients, on the other hand, cannot always predict the result. To address the issue, a suggested strategy employing the long short-term memory coordinated normal (LSTM) predicts the number of coronavirus cases in the coming 30 days as well as the effects of the virus.