Volume 18, No. 5, 2021
Covid-19 Prediction Modeling Using Bidirectional Gated Recurrent Unit Network Model
G.Tirupati , MHM Krishna Prasad , P.Srinivasa Rao
Abstract
Artificial Neural Networks (ANN) are Non-linear models can solve many complex and sophisticated real world problems. Deep Learning (DL) based forecasting mechanism has a significant role to enhance the future course of action in numerous domains. A variant Recurrent Neural Networks are being effectively applied to handle the problems concerned with prediction. In this paper, a Bidirectional Gated Recurrent Unit (BGRU) Neural Network model has been proposed to predict the total number of Confirmed, Deaths and Cured cases on SARS-CoV-2 (COVID-19) pandemic. Our model performance is determined by Mean Absolute Percentage Error (MAPE), R2 score, R2Adjusted score and Root Mean Square Error (RMSE). Finally, the BGRU is contrasted with Machine Learning Methods such as simple Linear Regression (LR), Least Absolute Shrinkage Selection Operator (LASSO), Support Vector Regression (SVR) and simple Exponential Smoothing (ES). The findings showthat BGRU outperforms among the existing Techniques.
Pages: 15-41
Keywords: COVID-19; Prediction; Deep Learning; Regression;Mean Absolute Percentage Error