Applying Machine Learning Prediction Methods to COVID-19 Data

Abstract

The Coronavirus (COVID-19) epidemic emerged in China and has caused many problems such as loss of life, and deterioration of social and economic structure. Thus, understanding and predicting the course of the epidemic is very important. In this study, SEIR model and machine learning methods LSTM and SVM were used to predict the values of Susceptible, Exposed, Infected, and Recovered for COVID-19. For this purpose, COVID-19 data of Egypt and South Korea provided by John Hopkins University were used. The results of the methods were compared by using MAPE. Total 79% of MAPE were between 0-10. The comparisons show that although LSTM provided the better results, the results of all three methods were successful in predicting the number of cases, the number of patients who died, the peaks and dimensions of the epidemic.

Time Series Cleaning Methods for Hospital Emergency Admissions

“…Box-Jenkins (ARIMA-Autoregressive Integrated Moving Average) models (AR-Autoregression, MA-Moving average, ARMA-Autoregressive moving average) [13] have been used in many fields such as furniture [14], finance [15], energy [16], food [17] for discrete and linear time series datasets. In the healthcare field, in addition to the emergency department density estimation [18] [19], covid-19 [20], the number of calls to the 112emergency call center [21], the average cost per prescription [22], the need for medical supplies [23], serum set consumption [24], Electrocardiogram (ECG) signal analyzes [25] and hospital disaster preparedness [26] have also been used for estimation purposes. LSTM (Long Short Time Memory) with recent success in deep learning approaches [27] [28] [29] has been used in many fields such as [30] financial [31], energy [32], health [33] [34] [35] [36].…”