Covid-19 pandemic prediction using artificial intelligence techniques in Türkiye

The COVID-19 pandemic has been a serious threat to people's lives since the first case was seen. The rapid spread of the virus around the world and the fact that it creates a global problem reveals the importance of the authorities' making quick and effective decisions. In this respect, detecting the COVID-19 disease and estimating the increasing number of cases has become an important issue. In this study, the number of daily cases, daily deaths, cumulative cases, and cumulative deaths are predicted in TURKEY. Stacked-LSTM, Bi-LSTM, and ANN methods, which are among the artificial intelligence techniques, are used for estimating the number of COVID-19 cases. The hyperparameters of these methods are optimized using the Grey Wolf Algorithm (GWO). In the study conducted in Turkey, MAPE, MAE, and R^2 Score performance metrics were used to determine the accuracy between the actual data and the predicted data for COVID-19. The success of the models was evaluated by comparing the datasets used for the models separately. With this study, the model and data set that give the closest result to the actual data values were determined. Thus, the accuracy of the pandemic was evaluated for the models.