Treffer: Forecasting the Number of COVID-19 Active Cases in West Java Using the Multilayer Perceptron Feedforwards Neural Networks.

West Java reported that as of July 17, 2021, 508,814 persons had been proven positive for COVID-19. On the same day, there were 115,257 active COVID-19 cases in West Java. In West Java, there will be a large increase in active COVID-19 cases in 2021. The number of active cases increased by 95,532 between June 5 and July 17, 2021. During that time, active cases grew by 484 percent. On July 17, 2021, the Bed Occupancy Ratio (BOR) in West Java was at 80.48 percent, exceeding the WHO standard of 60 percent. This has an effect on the number of patients who are rejected at the COVID-19 referral hospital. COVID-19 patients in need of medical attention are the focus of active cases. Active cases were predicted in this study using a Multilayer Perceptron (MLP). The COVID-19 Task Force provided the data for this study. The data show the number of COVID-19 positive cases, recovered, and deaths in 34 Indonesian provinces from 2 March 2020 to 17 July 2021. The results of the study found, from the results of the evaluation using data testing the number of active cases in the last 15 weeks, namely April 10 - July 17, 2021, the use of adam 0.1 provides relatively good accuracy compared to the use of other settings, namely with MAPE, RMSE and MAE values of 17.21%, 7092.25, and 5470. From 15 testing periods, MLP is accurate in predicting the number of active cases the first week 13 times, and the second week 9 times with an absolute percentage error (APE) <20%. It is hoped that the findings of this study would be useful to the government as a reference in conditioning hospital bed capacity to deal with active COVID-19 cases in West Java during the following two weeks, so that no COVID-19 patients are turned away because the hospital is full. [ABSTRACT FROM AUTHOR]

Copyright of Proceedings of the International Conference on Industrial Engineering & Operations Management is the property of IEOM Society International and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)