Time series prediction of the COVID-19 outbreak in India using LSTM based deep learning models
Proceedings of 2023 3rd International Conference on Innovative Practices in Technology and Management, ICIPTM 2023
; 2023.
Article
Dans Anglais
| Scopus | ID: covidwho-20241755
ABSTRACT
The epidemic caused by COVID-19 presents a significant risk to the continuation of human civilisation and has already done irreparable damage to society. In this paper, forecasting of Coronavirus outbreak in India is performed by LSTM and CovnLSTM deep neural network techniques. COVID-19 data of confirmed cases of India is used. It was taken from John Hopkins University. The loss rate of ConvLSTM is lower than LSTM and RMSE of ConvLSTM is lower than LSTM. For training Covn-LSTM shows 0.069% and testing ConvLSTM shows 0.32% improvement over LSTM model. Therefore, ConvLSTM outperformed over LSTM model. Further wise selection of hyper-parameters could increase the accuracy of the models. © 2023 IEEE.
Convolutional-Long Short term Mermory; Coronavirus; deep learning; forecasting; India; Convolutional neural networks; Deep neural networks; Long short-term memory; Coronaviruses; Human civilization; John Hopkins University; Learning models; Neural network techniques; Short term; Time series prediction; COVID-19
Texte intégral:
Disponible
Collection:
Bases de données des oragnisations internationales
Base de données:
Scopus
Type d'étude:
Études expérimentales
/
Étude pronostique
langue:
Anglais
Revue:
Proceedings of 2023 3rd International Conference on Innovative Practices in Technology and Management, ICIPTM 2023
Année:
2023
Type de document:
Article
Documents relatifs à ce sujet
MEDLINE
...
LILACS
LIS