An Improved Random Forest Algorithm for Predicting the COVID-19 Pandemic Patient Health

Abstract

The 2019 pandemic of the Corona infectious diseases also called COVID-19 has been emanated in the Asian country China in the year 2019. This is precariously unjustly preventing almost everything in human society. This same rapid pace as well as an incremental increase in the proportion of patient populations, consequently, required an efficient and accurate forecast of such an infected client's potential result for proper care utilizing the machine learning model. To overcome these needs, the research experiment strives to optimize Covid-19 patient's potential to estimate a revolutionary model focused on an improved random forest (IRF) methodology with safety various features. This proposed IRF method has been implemented to predict Covid-19 information about a persons condition with high-dimensional, unstable functionality. Initially, a random forest algorithm will be used for organizing the value of the dependent variable and reducing the measurements. By using the COVID-19 Kaggle online dataset set of data, the proposed IRF method as well as existing Random Forest (RF) and Support Vector Machines (SVM) has been implemented over the WEKA Machine learning simulation tool.