A Deep Learning Method to Forecast COVID-19 Outbreak

Dash, Satyabrata; Chakravarty, Sujata; Mohanty, Sachi Nandan; Pattanaik, Chinmaya Ranjan; Jain, Sarika

A new pandemic attack happened over the world in the last month of the year 2019 which disrupt the lifestyle of everyone around the globe. All the related research communities are trying to identify the behaviour of pandemic so that they can know when it ends but every time it makes them surprise by giving new values of different parameters. In this paper, support vector regression (SVR) and deep neural network method have been used to develop the prediction models. SVR employs the principle of a support vector machine that uses a function to estimate mapping from an input domain to real numbers on the basis of a training model and leads to a more accurate solution. The long short-term memory networks usually called LSTM, are a special kind of RNN, capable of learning long-term dependencies. And also is quite useful when the neural network needs to switch between remembering recent things, and things from a long time ago and it provides an accurate prediction to COVID-19. Therefore, in this study, SVR and LSTM techniques have been used to simulate the behaviour of this pandemic. Simulation results show that LSTM provides more realistic results in the Indian Scenario.