Machine Learning Approaches to Predict Covid-19 Severity

ABSTRACT

Machine learning has been successfully used in the medical field for the last few years. The emergence of covid-19 pandemic has seen researchers using machine learning approaches to detect and predict whether a patient has been infected by covid-19 or not. Cough and fevers are the two most likely symptoms of the covid-19. In this paper, the journals regarding the prediction of covid-19 severity based on symptoms have been discussed. Several researchers have established deep learning-based models for the prediction. They have used the test data from the hospitals for their research process. The KNN models, ANN models, and SVM models have been discussed. The limitations of the past research process have been evaluated to be the biased test data set and missing values. The methodology that was used for the research process has been described. Mixed method of data collection was used for the research processes. The secondary data was collected from reliable sources for this research process. Kaggle website was used to collect the test data regarding the covid-19 patients. The data analysis was done on the weka tool using various machine learning models. Copyright © 2022, Anka Publishers. All rights reserved.