An AI-based Analysis of the effect of COVID-19 Stringency Index on Infection rates: A case of India

Abstract

Purpose: The impact of the COVID-19 pandemic has already been felt worldwide, disrupting the unremarkable life of individuals. Social consequences and viral transmission are challenges that must be resolved to effectively overcome the problems that occur throughout this pandemic. The COVID-19 infection data about India were represented using different statistical models. In this paper, the authors focus on the data collected between 1^st January 2020 and 12^th April 2021, try analyzing the different indexes related to India, and predict the number of infected people in the near future. Based on the infection rate, it is possible to classify a country as “fixed,” “evolving” and “exponential.” Based on the prediction, some recommendations are proposed to contain the outbreak of the disease. This will also help the government and policymakers to identify and analyze various risks associated with 'opening up' and 'shutting down' in response to the outbreak of the disease. With the help of these models, it is possible to predict the number of cases in the near future.

Methodology:    COVID-19 Stringency Index, Government Response Index, and Containment Health Index calculated, published, and updated real-time by a research group from Oxford University (https://www.bsg.ox.ac.uk/research/research-projects/covid-19-government-response-tracker) on 21 mitigation and suppression measures employed by different countries were analyzed using a few mathematical models to find the relationship between Stringency Index and infection rates and forecast trends. A new model was proposed after analyzing a few mathematical models proposed by the researchers. Data analytics was also conducted using AI-based data analytics tools available online.  The dataset was kept updated until the date April 20, 2021, was downloaded for this purpose. The appropriate values were extracted from the original dataset and used to construct a sub-dataset, which was then used for the analytics. An AI-based online Data Analytics tool provided by datapine was used to forecast trends.

Findings/Result: It was observed that in India, as in other countries, there is a close association between Stringency Level and COVID-19 cases. The higher the degree of stringency, the lower the cases, and vice versa. The same can be said about the government's role and degree of containment & health.

Originality: In this paper, we analyzed various mathematical models for predicting the total number of COVID-19 cases and deaths due to COVID-19 in India. We also examined the relationship between total cases and the Government's Response Index, Containment & Health Index, and Stringency Index indicators. The model we proposed to predict COVID-19 cases on a day-by-day basis had a 98 percent accuracy rate and a 2% error rate.

Paper Type: Analytical. With prerecorded datasets obtained from online resources, and data analysis was conducted using mathematical models and AI-based analytical tools.