(Un)Masked COVID-19 Trends from Social Media
Authors:
Asmit Kumar Singh,
Paras Mehan,
Divyanshu Sharma,
Rohan Pandey,
Tavpritesh Sethi,
Ponnurangam Kumaraguru
Abstract:
Wearing masks is a useful protection method against COVID-19, which has caused widespread economic and social impact worldwide. Across the globe, governments have put mandates for the use of face masks, which have received both positive and negative reaction. Online social media provides an exciting platform to study the use of masks and analyze underlying mask-wearing patterns. In this article, w…
▽ More
Wearing masks is a useful protection method against COVID-19, which has caused widespread economic and social impact worldwide. Across the globe, governments have put mandates for the use of face masks, which have received both positive and negative reaction. Online social media provides an exciting platform to study the use of masks and analyze underlying mask-wearing patterns. In this article, we analyze 2.04 million social media images for six US cities. An increase in masks worn in images is seen as the COVID-19 cases rose, particularly when their respective states imposed strict regulations. We also found a decrease in the posting of group pictures as stay-at-home laws were put into place. Furthermore, mask compliance in the Black Lives Matter protest was analyzed, eliciting that 40% of the people in group photos wore masks, and 45% of them wore the masks with a fit score of greater than 80%. We introduce two new datasets, VAriety MAsks - Classification (VAMA-C) and VAriety MAsks - Segmentation (VAMA-S), for mask detection and mask fit analysis tasks, respectively. For the analysis, we create two frameworks, face mask detector (for classifying masked and unmasked faces) and mask fit analyzer (a semantic segmentation based model to calculate a mask-fit score). The face mask detector achieved a classification accuracy of 98%, and the semantic segmentation model for the mask fit analyzer achieved an Intersection Over Union (IOU) score of 98%. We conclude that such a framework can be used to evaluate the effectiveness of such public health strategies using social media platforms in times of pandemic.
△ Less
Submitted 9 July, 2021; v1 submitted 30 October, 2020;
originally announced November 2020.
Catching up with trends: The changing landscape of political discussions on twitter in 2014 and 2019
Authors:
Avinash Tulasi,
Kanay Gupta,
Omkar Gurjar,
Sathvik Sanjeev Buggana,
Paras Mehan,
Arun Balaji Buduru,
Ponnurangam Kumaraguru
Abstract:
The advent of 4G increased the usage of internet in India, which took a huge number of discussions online. Online Social Networks (OSNs) are the center of these discussions. During elections, political discussions constitute a significant portion of the trending topics on these networks. Politicians and political parties catch up with these trends, and social media then becomes a part of their pub…
▽ More
The advent of 4G increased the usage of internet in India, which took a huge number of discussions online. Online Social Networks (OSNs) are the center of these discussions. During elections, political discussions constitute a significant portion of the trending topics on these networks. Politicians and political parties catch up with these trends, and social media then becomes a part of their publicity agenda. We cannot ignore this trend in any election, be it the U.S, Germany, France, or India. Twitter is a major platform where we observe these trends. In this work, we examine the magnitude of political discussions on twitter by contrasting the platform usage on levels like gender, political party, and geography, in 2014 and 2019 Indian General Elections. In a further attempt to understand the strategies followed by political parties, we compare twitter usage by Bharatiya Janata Party (BJP) and Indian National Congress (INC) in 2019 General Elections in terms of how efficiently they make use of the platform. We specifically analyze the handles of politicians who emerged victorious. We then proceed to compare political handles held by frontmen of BJP and INC: Narendra Modi (@narendramodi) and Rahul Gandhi (@RahulGandhi) using parameters like "following", "tweeting habits", "sources used to tweet", along with text analysis of tweets. With this work, we also introduce a rich dataset covering a majority of tweets made during the election period in 2014 and 2019.
△ Less
Submitted 18 September, 2019; v1 submitted 16 September, 2019;
originally announced September 2019.
