Paper Key : IRJ************337
Author: Indumathi R
Date Published: 03 Mar 2023
Abstract
In this study, the number of novel coronavirus (COVID-19) positive reported cases for 32 Indian states and union territories is predicted using Deep Learning-based models. On an Indian dataset, the number of positive instances is predicted using recurrent neural network (RNN) based long-short term memory (LSTM) variations such as Deep LSTM, Convolutional LSTM, and Bi-directional LSTM. For forecasting daily and weekly cases, the LSTM model with the lowest error is used. It has been found that the suggested strategy produces short-term predictions with great accuracy, with error rates of less than 3% for predictions made each day and less than 8% for predictions made once a week. For the purpose of quickly identifying new coronavirus hotspots, Indian states are divided into various zones depending on the distribution of positive cases and daily growth rate.