COVID-19 as the disease of concern motivates various scientists to investigate it in various perspectives. In statistical perspective, a number of statistical models are used to predict the outcome of COVID-19 cases given a number of risk factors. Accuracy of a statistical model in predicting the outcome is important to be determined.

A part of supervised machine learning called deep learning is used to predict the outcome of COVID-19 given five predictors, new cases, age ≥ 65 years, prevalence of diabetes mellitus, female smoker, and male smoker. Big data of COVID-19 is downloaded from the website. A thousand data sets have been analyzed by neural network algorithm using library Keras.

Performance of neural network model in predicting COVID-19 outcome shows perfect accuracy with accuracy measure equal to 1.0000 and very small model error with loss equal to 3.2529e-08. Although the results achieve the goal, the process is time consuming which spends almost 23 hours. Neural network model is recommended to be used for predicting the outcome in small scale of big data. It is recommended to develop the algorithm which be less time consuming when large scale of big data having many variables as input and multi-layer perceptron is implemented.

accuracy loss neural, network deep learning.

Received: December 8, 2021; Accepted: January 19, 2022; Published: January 25, 2022

How to cite this article: Kuntoro Kuntoro, Accuracy of neural network model in predicting outcome of covid 19 using deep learning approach, JP Journal of Biostatistics 19 (2022), 113‑121. DOI: 10.17654/0973514322008

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License 

References:

[1] G. Casella and R. L. Berger, Statistical Inference, Second Edition, Australia, Duxbury Thomson Learning, 2002. https://github.com/owid/covid-19-data/blob/master/public/data/latest/owid-covid-latest.csv 121121 18.57.
[2] H. Kuntoro, Metode Statistik - Edisi Revisi. Surabaya, Pustaka Melati. 2011, pp. 16-17.
[3] P. S. Levy and S. Lemeshow, Sampling of Populations- Methods an Applications Fourth Edition, New York, A John Wiley & Sons, Inc., Publication, 2008, 35-36.
[4] W. S. McCulloch and W. Pitts, A logical calculus of the ideas immanent in nervous activity, Bulletin of Mathematical Biology 52(l/2) (1990), 99-115.
[5] P. C. Petersen, Neural Network Theory, 2020.
http://pc-petersen.eu/Neural_Network_Theory.pdf 251221 7.42pm.
[6] F. Rosenblatt, The Perceptron: A probabilistic model for information storage and organization in the Brain, Psychological Review 65(6) 1958.