Prediction for the spread of COVID-19 in India and effectiveness of preventive measures

• Published in: The Science of the total environment 2020-08, Vol.728, p.138762-138762, Article 138762
• Main Authors: Tomar, Anuradha, Gupta, Neeraj
• Format: Article
• Language: English
• Subjects: Betacoronavirus; Coronavirus Infections - epidemiology; Coronavirus Infections - prevention & control; COVID-19; Curve fitting; Forecasting; Humans; India - epidemiology; LSTM; Models, Theoretical; Neural Networks, Computer; Pandemics - prevention & control; Pneumonia, Viral - epidemiology; Pneumonia, Viral - prevention & control; Prediction; Recurrent neural network; SARS-CoV-2
• ISSN: 0048-9697; 1879-1026; 1879-1026
• DOI: 10.1016/j.scitotenv.2020.138762

The spread of COVID-19 in the whole world has put the humanity at risk. The resources of some of the largest economies are stressed out due to the large infectivity and transmissibility of this disease. Due to the growing magnitude of number of cases and its subsequent stress on the administration and health professionals, some prediction methods would be required to predict the number of cases in future. In this paper, we have used data-driven estimation methods like long short-term memory (LSTM) and curve fitting for prediction of the number of COVID-19 cases in India 30 days ahead and effect of preventive measures like social isolation and lockdown on the spread of COVID-19. The prediction of various parameters (number of positive cases, number of recovered cases, etc.) obtained by the proposed method is accurate within a certain range and will be a beneficial tool for administrators and health officials. [Display omitted] •Data-driven model based on LSTM techniques is proposed for COVID-19.•90 days day-ahead estimation for various parameters.•Analysis on effect of lockdown, social isolation and impact of transformation ratio.•Forecasting and an insight towards possible situation in coming days.