ABSTRACT
Nowadays, social media platforms generate an immense amount of information in the form of text, images, video, sound, among others. Their capabilities and reliability during adverse situations have made them society's go-to communication method as they continue to operate while more traditional methods fail [1]. With the unexpected arrival of the COVID-19 pandemic, billions of tweets had been generated, bringing both opportunities and challenges to emergency managers when seeking to leverage social media data as a source of information. Therefore, this research investigates how emergency managers could utilize social media data for monitoring public sentiment to enhance their strategic decision-making process. To achieve our end objective, we have adapted a visual analytics framework that has been developed for alerting and monitoring public sentiment during product recalls [2]. The proposed work understands that by developing an alert warning system based on collective sentiment analysis, decision makers will be able to identify scenarios where significant levels of negative or positive sentiment are being disseminated. The alert warning system framework includes concepts on data analytics, natural language processing, and machine learning techniques as mechanisms to generate inferences from social media applications. To illustrate our work, we extracted a sample of 24.7 millions of COVID-19 related tweets from the region of El Paso, TX, which in November 2020 was one of the most dangerous COVID-19 hotspots in the United States [3]. Our results indicate that the adapted framework is an initial step when seeking to assist emergency managers when seeking to utilize social media data;however, it has been found that additional challenges must be addressed before emergency domain decision makers can fully adopt it into their management strategies. © 2022 IEEE.