An Apprehensive Analysis of Twitter Data for Disaster Management: A Case Study of Hurricane Harvey

Abstract:

The use of social media platforms such as Twitter is significantly increasing during natural disasters. With the emergence of several social media platforms over the past decade, many studies have been conducted on their applications during calamities. This study presents a comprehensive analysis of textual content from millions of tweets shared on Twitter during the hurricane Harvey across the several counties in the southeast Texas where Harvey made landfall. Our analysis indicates a dramatic increase in tweet activity when the hurricane hit the study area. We utilized a dataset of 18 million tweets through categorizing them across the counties in the southeast Texas.

In this study, we employed multiple Artificial Intelligence techniques from Natural Language Processing (NLP), to perform analysis on the textual data generated by the twitter users during the hurricane Harvey. This enabled us to subdivide the tweet contents to several categories with the aim of benefiting crisis management during the event. Our study also provides a variety of useful information at county level before, during and after Harvey that can significantly help disaster managers and responders to minimize the consequences of the event and improve the preparedness of the residents against it.