Analysis and Comparison for Innovative Prediction Technique of COVID-19 using Logistic Regression algorithm over Support Vector Machine Algorithm with Improved Accuracy

ABSTRACT

Aim: The main objective of this study is to improve the accuracy of COVID-19 prediction and evaluation. Material(s) and Method(s) This work depends on the data segregated from Kaggle's website where the samples are divided into two groups. Each group contains 20 samples (N=20) for both the Logistic regression and Support vector machine algorithms in accordance with the total sample size calculated using clinicalc.com by keeping alpha error-threshold at 0.05, confidence interval at 95%, enrolment ratio as 01, and G power at 80%. This involves training the data with validating 20 validations ranging from 5 to 24 in MatLab 2021a. Result(s) The accuracy, sensitivity, and precision rates are compared using the SPSS Software and Independent sample T-test. The Logistic regression has better accuracy, sensitivity, and precision of 95.98%,94,65%, 96.20% (P<0.001) respectively compared to the Support vector machine where 91.25% of accuracy (P<0.001), 93.93% of sensitivity (P<0.001), and 86.11% of precision (P<0.001). Conclusion(s) The Logistic regression algorithm produces superior outcomes than the Support vector machine algorithm. Copyright © 2022 Wolters Kluwer Medknow Publications. All rights reserved.