Understanding Peoples' Sentiment During Different Phases of COVID-19 Lockdown in India: A Text Mining Approach

Understanding Peoples' Sentiment During Different Phases of COVID-19 Lockdown in India: A Text Mining Approach

Rabindra Ku Jena, Rupashree Goswami
Copyright: © 2021 |Pages: 17
DOI: 10.4018/IJBAN.2021100104
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

During a global pandemic like COVID-19, the success of governmental policies depends on the people's sentiments and extended cooperation towards these policies. Therefore, this study explores the prevalent discourse in social media about different aspects of the COVID-19 pandemic and the policies to manage and control it. Data from Twitter collected between 25 March 2020 and 1 July 2020 was used for topic modelling and sentiment analysis. Natural language processing-based text mining techniques were used for analysis. This study first identified different frequent COVID-19-related topics and then analyzed how the sentiments towards these topics differ across different phases of lockdown. Further, insights into how different topics were perceived by gender and age group are also discussed in this study. Finally, this study also analyzed how daily casualty due to COVID-19 influenced the public sentiments and number of daily tweets. The study provides a robust NLP-based text mining framework to predict the people's sentiment during COVID-19 lockdown in India. The insights presented in this study can help the authorities mitigate the COVID-19 pandemic effectively and help different agencies in the face of similar pandemics in the future.
Article Preview
Top

1. Introduction

The world witnessed the Corona Virus Disease (COVID-19) in December 2019, which originated from the Wuhan provinces of the Republic of China. Due to the spread of COVID-19, countries worldwide are facing unusual challenges in managing and controlling the epidemic. Government authorities worldwide are pursuing to slow down the spread of COVID-19 and sustain healthcare and other related facilities. Significant initiatives such as home isolation, home quarantine, social distancing, and closure of public amenities at highly affected regions are among the top priorities of the authorities. These initiatives have been taken to minimize the community spread of COVID-19. Recent studies have also found that the policies like social distancing, isolations can effectively suppress the virus spread (Kucharski et al., 2020; Ferguson et al., 2020). The government of India has taken various public health measures and series of lockdown to control the spread of COVID-19 after the detection of the first confirmed case on 30 January 2020 (MyGOV, 2020). The unprecedented lockdown measures were aimed to contain the widespread disease. The effectiveness of all these steps depends on public awareness, continuous public support, and durable compliance. Figure 1 showed the information about different lockdown phases in India since the first case was reported on 30 January 2020. The figure also showed an incremental growth of both COVID-19 cases and casualties in different phases.

On the other hand, in a democratic country like India, people react differently to various steps taken by the government to tackle and manage the spread of COVID-19. Hence, understanding people's perceptions are essential to implement effectively and improve the measures taken by the government during different lockdown phases. Detailed analysis of Twitter data can provide critical information about people's perceptions towards different aspects of COVID-19 lockdown.

Figure 1.

Progress of COVID-19 Pandemic and fear curve

IJBAN.2021100104.f01

Nowadays, social media platforms play a vital role in capturing and disseminating peoples' perceptions of each and everything globally (Marino et al., 2018; Shwartz-Asher et al., 2017). Sentiment analysis techniques using social media platforms are being used extensively as an efficient tool to analyze, interpret, and classify the public perception towards any given issue (D'Avanzo et al., 2017; Pang & Lee, 2008). In the era of social media, sentiment analysis can efficiently analyze the textual data to gauge people's opinion. Social media platforms, mainly Twitter, have been used extensively to monitor and predict public sentiment and emotions (Jordan et al., 2018). Knowing the importance of public perception for the success of governments' policies to curb any pandemic, this study's primary objective is to assess the Indian peoples' sentiment towards COVID-19 preventive lockdown measures through sentiment analysis.

The rest of the paper is structured as follows: Section 2 reviewed the related research in sentiment analysis. Section 3 discusses the research methodologies adopted in this study. The detailed results and discussion are presented in section 4. Finally, conclusions, limitations and directions of future research are discussed in section 5.

Top

2. Literature Review

The psychological state of people plays a vital role in reacting to a pandemic. Understanding people's emotional state helps the stakeholders frame guidelines and take essential measures to manage the situation effectively. Recently several studies have been conducted to analyze the psychological state of people during COVID-19. On the other hand, sentiment analysis supports understanding users' behaviour and perception of the events worldwide (Lytras et al., 2020, Jena, 2019; Adinolfi et al., 2017; D'Avanzo et al., 2017). Various studies have been conducted to discover peoples' sentiments and emotions in different domains and topics of interest. Many studies have been conducted to identify the perceptions, sentiments and emotions about different aspects of the COVID-19 pandemic in different parts of the world. Some of the key study findings are presented in this section.

Complete Article List

Search this Journal:
Reset
Volume 11: 1 Issue (2024)
Volume 10: 1 Issue (2023)
Volume 9: 6 Issues (2022): 4 Released, 2 Forthcoming
Volume 8: 4 Issues (2021)
Volume 7: 4 Issues (2020)
Volume 6: 4 Issues (2019)
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 4 Issues (2015)
Volume 1: 4 Issues (2014)
View Complete Journal Contents Listing