COVID-19 Future Forecasting Using Supervised Machine Learning Models
Abstract: AI (ML) based estimating systems have demonstrated their importance to expect in perioperative results to further develop the decision making on the future course of activities. The ML models have for some time been utilized in numerous application spaces which required the distinguishing proof and prioritization of unfavorable variables for a danger. A few expectation techniques are in effect famously used to deal with estimating issues. This study shows the capacity of ML models to gauge the quantity of impending patients impacted by Coronavirus which is by and by considered as a possible danger to humankind. Specifically, four standard guaging models, like straight relapse (LR), least outright shrinkage and choice administrator (Tether), support vector machine (SVM), and outstanding smoothing (ES) have been utilized in this review to gauge the compromising elements of Coronavirus. Three kinds of expectations are made by every one of the models, like the quantity of recently contaminated cases, the quantity of passings, and the quantity of recuperations in the following 10 days. The outcomes created by the review demonstrates it a promising system to involve these strategies for the current situation of the Coronavirus pandemic. The outcomes demonstrate that the ES performs best among every one of the pre-owned models followed by LR and Tether which performs well in anticipating the new affirmed cases, demise rate as well as recuperation rate, while SVM performs ineffectively in all the forecast situations given the accessible dataset.
Keywords: Exponential Smoothing, Least Absolute Shrinkage and Selection Operation, Linear Regression, Support Vector Machine
How to Cite?: Dhahabiyya Asharef, "COVID-19 Future Forecasting Using Supervised Machine Learning Models", Volume 11 Issue 2, February 2022, International Journal of Science and Research (IJSR), Pages: 960-967, https://www.ijsr.net/getabstract.php?paperid=SR22221205524, DOI: https://dx.doi.org/10.21275/SR22221205524