Title: Modified COVID-19 Indian and international dataset for automatic prediction of risk in an individual using machine learning models using a mobile APP
Authors: Jatin Bindra; Savita Ahlawat; Mohammed Javed
Addresses: Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi-110058, India ' Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi-110058, India ' Department of Information Technology, Indian Institute of Information Technology, Uttar Pradesh-211015, India
Abstract: COVID-19 is a major problem not only impacting the health but also the economic development of countries around the globe. Therefore, automatic COVID-19 diagnosis and risk prediction in an individual is very significant in preventing pandemic and also for proper medication. This paper adopts five machine learning models intended to predict the risk of having COVID-19. The existing datasets- the Indian COVID-19 dataset and the International COVID-19 Dataset have been systematically modified to include negative COVID-19 patient data and also facilitate feature learning. The modified datasets are experimented with models like logistic regression, Naive Bayes, k-nearest neighbours, random forest, and neural network. Further, a comparison is done in these models on the basis of score obtained in prediction on the two datasets. The whole model is also implemented as a Mobile APP to be used in real-time.
Keywords: modified COVID-19 dataset; machine learning; neural network; random forest; logistic regression; Naive Bayes; k-nearest neighbours; COVID-19 mobile APP.
International Journal of Intelligent Engineering Informatics, 2021 Vol.9 No.2, pp.142 - 160
Received: 01 Dec 2020
Accepted: 26 Jan 2021
Published online: 29 Jul 2021 *