Article Preview
TopIntroduction
The support vector machine (SVM) algorithm has showed high efficiency in solving classification issues in medical fields. SVM is data-driven and model-free that is discriminative for prediction in particular cases where sample sizes are small. This technique has been used to improve methods for predicting disease in the clinical setting (Battineni, G., et al., 2019). Support Vector Machine is a discriminative classifier that can be defined by a separating hyper-plane. It is the generalization of maximal margin classifier which comes with the definition of hyper-plane (Islam, M., 2017). SVM is characterized by an optimal hyper-plane to classify new examples and datasets (Gholami R & Fakhari N, 2017).
As COVID-19 is declared as an international epidemic in mid of March 2020 and more than 125000 confirmed cases have been recorded around the world, so day level forecasting about COVID -19 spread is crucial to measure the behavior of this new virus globally (WHO, 2020).
Corona virus is spreading around the world, causing panic to all men, for regional distribution of COVID-19 cases worldwide see Figure 1. It causes serious breathing-related symptoms particularly for elderly people and those suffering from chronic illnesses. For the past 6 months, diagnosing Corona virus disease is too complicated as it takes a long time to get results for COVID-19 tests based on the signs and symptoms. There is still no approved antiviral drug for treating COVID-19 until now.
Figure 1. Geographic distribution of COVID-19 cases worldwide, as of 16 August 2020 (ecdc)
From figure 1, the total numbers of infected cases of corona virus over the world is constantly increased during the past six months.
In Egypt, according to the Egyptian ministry of health, the infected cases of Corona virus are shown in figure 2. And the death cases of Corona virus are shown in figure 3.
Figure 2.
Total Corona virus Cases in Egypt, as of 16 August 2020 (worldometers)
Figure 3.
Total Corona virus death Cases in Egypt, as of 16 August 2020 (worldometers)
From figures 2 and 3, we notice the total numbers of corona virus in Egypt start with low infected and death cases during the first three months from mid-February to mid-May then total number of infected and death cases increase quickly with high numbers from mid-May to late July then total numbers start to decrease during August.
Corona virus causes a real health crisis around the world, beside its social, economic and political effects, and to help in solving this problem and avoiding infection and planning the healthcare system for possible potential up-comings, there is an urgent need to construct forecasting model providing future forecasts of the possible number of daily cases and daily deaths. In some case The accuracy of traditional forecasting is not high because it depends primarily on the availability of data but, in disease outbreaks such as current corona virus outbreak, there are no data at all at first and then limited as time goes by, there are concerns that the data may not be reliable.