Weighted Fuzzy Time Series Model to Forecast Epidemic Injuries and its Data Visualization | Chapter 5 | Current Overview on Science and Technology Research Vol. 5

 Globally, the coronavirus disease spread rapidly. To aid in decision-making, we must develop precise future projections of the number of COVID-19 infections. On the basis of accumulating global experience from which we are working to enhance our future methodologies to deal with such a pandemic and in an effort to find the best application, we present some of the techniques that have shown useful in predicting the number of people infected with COVID-19 in a specific period in different regions of the world in this paper.

We use a rewarding model to predict injuries in areas with COVID-19, particularly in the Arab region.

This prediction is based on Saudi Arabia's epidemic injury statistics from March 2 to July 20, 2020. Charts and graph visualisation are two alternative techniques to present time series data. We propose the use of weighted fuzzy time series methods (WFTS) and weighted non-stationary fuzzy time series techniques to compare with the conventional Auto-Regressive Integrated Moving Average (ARIMA) statistical method (WNSFTS). Before utilising the (ARIMA) and (WFTS) algorithms to forecast the given data, stationary data conversion is required because the data is not stationary. On our injuries dataset, we log transform and differentiate. The graphic depicts the situation for COVID in Saudi Arabia as improving. When we analyse the original data using the Dickey-Fuller Test (DFT), we find that the p-value is equal to 0.646, which is higher than 0.05 and shows non-stationarity. The mean square error (MSE), root mean square error (RMSE), and normalisation root mean square error (NRMSE) are used to compare the accuracy of the different techniques. The results show that WFTS approaches offer reliable assistance for calculating COVID-19-based pandemic injuries in the area. The use of Weighted Non Stationary Fuzzy Time Series (WNSFTS) can produce significantly better results when projecting the issue of epidemic injuries. because it can anticipate future infections and has a high level of predictive accuracy.

Author(s) Details:

Hala Ahmed Abdul-Moneim,
Department of Mathematics, Darb University College, Jazan University, Jazan, Saudi Arabia and Department of Mathematics, Faculty of Science, Minia University, Minia, Egypt.

Please see the link here: https://stm.bookpi.org/COSTR-V5/article/view/8335