id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-315676-y0qbkszx	Shahid, Farah	Predictions for COVID-19 with Deep Learning Models of LSTM, GRU and Bi-LSTM	2020-08-19		.txt	text/plain	2794	163	55	In this paper, proposed forecast models comprising autoregressive integrated moving average (ARIMA), support vector regression (SVR), long shot term memory (LSTM), bidirectional long short term memory (Bi-LSTM) are assessed for time series prediction of confirmed cases, deaths and recoveries in ten major countries affected due to COVID-19.  Statistical models as ARIMA, ML technique of SVR with polynomial and RBF kernels, and DL mechanisms of LSTM, GRU and Bi-LSTM are proposed to predict the COVID-19 three categories, confirmed cases, deaths and recovered cases for ten countries. Parameters with their values of SVR, ARIMA and LSTM is shown in Table 1 , while results of actual and predicted cases in three categories in terms of performance measures are presented in Table 2 .  COVID-19 dataset has been modelled using various regressors including ARIMA, SVR with polynomial and RBF kernels, LSTM, GRU and Bi-LSTM for future predictions on confirmed cases, deaths and recovered case for ten countries across the globe.	./cache/cord-315676-y0qbkszx.txt	./txt/cord-315676-y0qbkszx.txt