Study the Impact of COVID-19 on Twitter Users with respect to Social Isolation

Abstract

The COVID-19 pandemic is a major public health issue that has affected millions of people in many different manners. This research proposes a novel approach to study and monitor this pandemic's impact on Twitter users concerning Social Isolation. Various sets of analyses is conducted to identify potential correlations between tweets related to social isolation, people's movement trends, and their emotional status. In this research, multiple datasets are used, ranging from our own collected dataset to a huge COVID-19 twitter dataset with over 140 million tweets and from the Google mobility report to the World Health Organization dataset. A couple of cloud-based services are used to analyze this data and extract patterns and insights from them. We believe the results of this research can be used as a public health indicator to anticipate the possibility of social isolation and design the health policies accordingly.

A Large-Scale COVID-19 Twitter Chatter Dataset for Open Scientific Research—An International Collaboration

“…With a wide range of applications, it is vital for such resources to be published for maximum exposure and to benefit additional researchers in communities that might not have already adopted the use of Social Media data. We showed the impact this work has had on other researchers' work and the potential it has for epidemiological researchers [25,26] as well as other communities that have directly or indirectly performed epidemiologically related research during the pandemic [9,25,30,32,33,35,[37][38][39][40][41][42]. Applications related to extracting patient narratives to understand the disease progression of COVID-19 (and long COVID) from the patient perspective [43,44] are one vital product of resources such as this one.…”

Potential Pitfalls With Automatic Sentiment Analysis: The Example of Queerphobic Bias

“…A search of the SCOPUS 3 and ASSIA 4 databases for 'sentiment analysis' shows that automated tools are widely used in the social sciences to conduct research on topics as diverse as understanding the impact of COVID and other disasters (Kaur et al, 2020;Razavi & Rahbari, 2020); to analysing the online behaviours of vulnerable groups (Saifee et al, 2020). Much recent work relies on lexicon-based approaches such as LIWC (Tausczik & Pennebaker, 2010), VADER (Hutto & Gilbert, 2014) and SentiStrength (Thelwall et al, 2010), likely due to their ease of deployment.…”

“…It has been predicted that use of ML will become the norm in social science research (Robila & Robila, 2020). Examples of papers already using ML approaches include Kaur et al (2020) and Li et al (2020). Deep learning models can offer more sophisticated analysis having learned abstract patterns from vast amounts of data.…”

The Impact of COVID-19 in Depression in Arab Twitter Users

“…The COVID-19 pandemic is posing very serious public health challenges worldwide [1], with the COVID-19 pandemic having an impact the world has never seen before. The COVID-19 pandemic An International Journal of Computational Intelligence Methods, and Applications Areej Alzoubi, Ahmad Alaiad, Khaled Alkhattib, Ahed J Alkhatib, Aseel Abu Aqoulah, Mohmad Alharoun, and Almo'men bellah Alawnah, The Impact of COVID-19 in Depression in Arab Twitter Users has had a significant impact on the world as a whole in all areas of life, such as health and education, as the COVID-19 pandemic has had a clear and significant impact on millions of people around the world [2]. During the COVID-19 pandemic, some vulnerabilities in countries around the world that were not previously exposed, these effects have caused heavy and costly burdens on everyone and caused feelings of fear, anxiety and depression, so there is an urgent need to increase attention to providing effective support to the population [3].…”