Comparison of Supervised Learning Models for COVID-19 Confirmed Cases Prediction using Correlation Analysis
Jun‐Su Kim, Byung-Jae Choi · 2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · 2022
The COVID-19 virus, which first appeared in 2019, has a strong contagious power and is highly spread by people’s mobility. In this study, correlation analysis is used in statistical preprocessing of dataset which further used to predict the COVID-19 confirmed cases for next day. Data is divided into two sets by organizing the data set by data preprocessing using correlation analysis. The first dataset is Google Mobility Data of COVID-19 infection with six variables. The second dataset is Google Mobility Data of COVID-19 infection with two variables: (1) Retail stores and leisure facilities (2) Grocery stores and pharmacies. The results of predicting the number of confirmed cases are compared using four supervised machine learning models. Furthermore, the soft voting method is used to show more improved results than the individual performances of each method.