CUR - The Council on Undergraduate Research

Visualization of Sentiments for Covid-19 Mask Usage

Eric Landaverde, Shivam Patel, Dr Jean Chu, Dr. Daehan Kwak School of Computer Science and Technology, Kean University, Union, NJ 07083

Covid-19 continues to leave its mark on the United States as the number of cases continues to rise with no foreseeable end, and what can be done to contain the pandemic and prevent it from leaving long term damage has become increasingly unclear. This confusion around covid-19 and the sentiments around proposed solutions warrant investigation; thus, the objective of this study is to examine the sentiments around one of the proposed measures to contain the spread of covid-19; mask usage. This study examines the sentiments of twitter users on mask usage and visualizes them on a map to locate hotspots of positive and negative perceptions and investigate its relation to spikes or drops in cases. The collection and visualization of this data will serve to inform professionals and institutions on how to tackle these perceptions to increase the adoption rate of preventative measures like masks. This will be done through a tweet mining system designed to scrape tweets using a keyword list and then obtain a user’s bio data and geo-location to locate the state, county, or city of a user. Visual analysis will be conducted on the dataset using numerous sentiment analysis APIs (e.g IBM Watson) to visualize the sentiments on a map using the Google Maps API to show hotspots of positive and negative sentiments in relation to the number of covid-19 cases. Negative and positive sentiments toward mask usage is not expected to correlate with the case totals because of factors like social pressure that result in actions diverging from what is said or tweeted. The goal is to gain an understanding of how divided people are on mask usage and perceptions of mask usage, and how they relate to the number of cases in an area. 

Additional Abstract Information