Classification Analysis of COVID19 Patient Data at Government Hospital of Banyumas using Machine Learning

Abstract

The development of the COVID-19 pandemic has not ended for almost 2 years. Even new variants appear that are more worrying. Including cases of COVID-19 in the Banyumas Raya area, a new variant from India entered through the Cilacap district. The objective study is to analyze the classification of COVID-19 patient data at the Government Hospital (RSUD) Banyumas from December 2020 to March 2021. In this analysis, we use several Machine Learning (ML) algorithms, including Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighborhood (KNN), Naïve Bayes, and linear regression. The variable used are vital sign factor which are blood pressure, temperature, Respiratory Rate (RR), SpO2, pulse rate, age, and age category. The class variable is age category. Based on the data obtained, a number of 6,464 patients are categorized as elderly. In general, the vital sign examinations show that they are within normal limits, except for the rate of respiration (RR), which is an average of 21 cycles per minute, which should normally be 8-12 cycles per minute. The classification process of age category variables shows that the RF algorithm provides the highest classification accuracy of 99.92%. For the future, this dataset could be examined by using Deep Learning (DL) algortihms to improve the accuracy.