Moreover, the details of these work are summarized in Table 1. For more details about this topic, the interested readers can refer to 17, 18, 19, 20, 21, 22, 23. The Chaos, Solitons and Fractals launched an open focus issue for understanding and mitigating the effects of the current pandemic 16. 15 used grey models to study the number of patients infected with COVID-19. Castillo and Melin 12 proposed a hybrid intelligent fuzzy fractal method for COVID-19 classification of countries. Sun and Wang 11 examined the data from January 23 to March 25 by ordinary differential equation model, which demonstrate that strongly controlled measured can minimize total infections. 10 presented a multiple ensemble neural network with fuzzy logic method for the COVID-19 cases in Mexico where the errors are significantly lower than traditional neural networks. 9 estimated the number of COVID-19 epidemic cases of Turkey, Germany, United Kingdom, France, Italy, Russia, Canada and Japan by Box-Jenkins (ARIMA), curve estimation models and Brown/Holt linear exponential smoothing methods. Hawas 8 introduced the recurrent neural networks for forecasting the virus’s daily infection in Brazil with limited raw data. 7 developed a deep learning method with rolling mechanism to forecast the epidemic trend for Russia, Peru and Iran. 6 used neural network with Stacked LSTM, Convolutional LSTM and Bi-directional LSTM to study the confirmed cases and the death cases of COVID-19 in USA and India. Petropoulos and Makridakis 5 introduced an objective method to predict the spread of confirmed cases, the number of deaths and recoveries of the COVID-19 under the assumption that the original data is reliable and the process of the disease following the past pattern. 4 used a Susceptible-Infectious-Recovered-Dead (SIDR) model to study the basic reproduction number, the per day infection mortality and the recovery rates of Hubei in China.
The possible trends and stopping time of COVID-19 in Canada are evaluated, and then compared transmission rates of Canada with Italy and USA. Chimmula and Zhang 3 proposed a new state-of-the-art Deep Learning forecasting model for COVID-19 outbreak in Canada. Publicly available datasets of 10 countries are used to establish the fuzzy model, and the results show the new model can be considered good studying the complexity of this epidemic diseases.
QUADRATIC POLYNOMIAL SERIES
Castillo and Melin 2 described a hybrid intelligent approach for efficient and accurate prediction COVID-19 time series combining fuzzy logic and fractal theory. It is generally known that the statistical models like autoregressive model, moving average and autoregressive integrated moving average, and the computational intelligence methods are widely applied in COVID-19 diseases. Thus accurately prediction the tendency, particularly at the early stage of the disease, can give a guidance for the control and prevention of the coronavirus. Moreover, there is no indication that the virus will disappear within a few months. At now, the total confirmed cases has reached 137,866,311 cases all over the world. The number of confirmed cases rose sharply since the January 2020, and governments have to promulgate various laws and policies to alleviate the spread of COVID-19. It is proven that this coronavirus can be transmitted from person to person. This disease can lead to severe fever, and mainly acute respiratory failure syndrome 1. At the beginning of 2020, a new strain of coronavirus (COVID-19) was found from some patients in January 2020.
0 kommentar(er)
