COVID-19 Case Number Prediction Utilizing Dynamic Clustering With Polynomial Regression
Felix Zhan · 2021
The novel coronavirus disease 2019 (COVID-19) has spread rapidly throughout the world since its first reported case in 2019. As of December 2020, over 70 million cases and 1.5 million deaths have been reported [3]. As a result, the pandemic has sparked a new, widespread need for knowledge discovery and prediction using the available data. In this project, I applied Dynamic Clustering with Polynomial Regression (DyCPR) to a dataset of daily increases in Coronavirus cases. The results show that for some US states, DyCPR performs better than Moving Average in predicting the number of new COVID-19 cases on the day following a limited subsequence of the time-series data.