A Comparative Study on Covid-19 Cases in Top 10 States/UTs of India in Using Machine Learning Models

Main Article Content

Dr. Deena Babu Mandru, et. al.


Coronavirus is a dangerous sickness came from a new virus. It has been assumed as an overall pandemic and a very hard circumstance to control the COVID-19 epidemic in India and global and so needed some severe actions to control its rate of increment. This disease causes for cold, dry cough, high fever, sore throat and serious breathing problems. This paper presents analysis of confirmed, cured and deaths cases, age and gender based cases in top 10 States/UTs of India. We analyzed various trends and patterns from various state/UTs units, MHFW of India data sources (up to 16th November 2020). Now a day’s plentiful models are proposed to predict covid-19 cases in India and world countries. The novel COVID-19 datasets are taken from Kaggle and GitHub repositories to analyze the epidemiological cases of the disease in top 10 states/UTs of India. We used various machine      learning algorithms like Linear Regression, KNN Regressor, LASSO Regression, Elasticnet Regression and Decision Tree regressor to analyze the number of novel Coronavirus (COVID-19) reported cases in top 10 states/UTs of India. The model analyzes datasets containing the COVID-19 cases (confirmed, cured and death cases) up to 24th November, 2020 using ML models. From the results it is proven that Decision Tree and KNN regressor performs best in analyzing the number of confirmed cases and number of death cases. But for number of cured cases LASSO and linear regression models give the best accuracy results. Unfortunately, Elastic net produced poor accuracy results due to some changes in original datasets. Especially, this work analyzes the calculations based on the exactness rate on a test dataset.


Download data is not yet available.


Metrics Loading ...

Article Details

How to Cite
et. al., D. D. B. M. . (2021). A Comparative Study on Covid-19 Cases in Top 10 States/UTs of India in Using Machine Learning Models . Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(10), 4514–4524. Retrieved from https://turcomat.org/index.php/turkbilmat/article/view/5194