Analyzing the Death Ratio of Covid Patients using Multiple Logistic Regression in Comparison with Lasso Regression for Improving Accuracy
DOI:
https://doi.org/10.47750/pnr.2022.13.S04.003Keywords:
Big Data, Supervised Learning, Death Ratio, Lasso Regression, Novel Multiple Logistic Regression, Machine Learning.Abstract
Aim: The idea of this study is to analyze and improve the death ratio accuracy of covid patients with Novel Multiple Logistic Regression(MLR)and Lasso regression. Both these algorithms fall under supervised learning techniques.
Materials and Method: Accuracy is analyzed for covid dataset of size 239 places. Analyzingthe death ratio of covid patients is performed by a Novel Multiple Logistic Regression of sample size (N=35) and Lasso regression of sample size (N=35), obtained using the Gpower value 80%. These are Supervised learning algorithms.
Result: Novel Multiple Logistic Regression accuracy is 96% which is comparatively higher than LAS with accuracy of 66%. The significance is determined as p=0.029 (p<0.05) for obtaining accuracy.
Conclusion: NovelMultiple Logistic Regression performs better in determining accuracy than Lasso Regression.
