Crisis Communication Patterns in Social Media during Hurricane Sandy

📝 Abstract

Hurricane Sandy was one of the deadliest and costliest of hurricanes over the past few decades. Many states experienced significant power outage, however many people used social media to communicate while having limited or no access to traditional information sources. In this study, we explored the evolution of various communication patterns using machine learning techniques and determined user concerns that emerged over the course of Hurricane Sandy. The original data included ~52M tweets coming from ~13M users between October 14, 2012 and November 12, 2012. We run topic model on ~763K tweets from top 4,029 most frequent users who tweeted about Sandy at least 100 times. We identified 250 well-defined communication patterns based on perplexity. Conversations of most frequent and relevant users indicate the evolution of numerous storm-phase (warning, response, and recovery) specific topics. People were also concerned about storm location and time, media coverage, and activities of political leaders and celebrities. We also present each relevant keyword that contributed to one particular pattern of user concerns. Such keywords would be particularly meaningful in targeted information spreading and effective crisis communication in similar major disasters. Each of these words can also be helpful for efficient hash-tagging to reach target audience as needed via social media. The pattern recognition approach of this study can be used in identifying real time user needs in future crises.

💡 Analysis

Hurricane Sandy was one of the deadliest and costliest of hurricanes over the past few decades. Many states experienced significant power outage, however many people used social media to communicate while having limited or no access to traditional information sources. In this study, we explored the evolution of various communication patterns using machine learning techniques and determined user concerns that emerged over the course of Hurricane Sandy. The original data included ~52M tweets coming from ~13M users between October 14, 2012 and November 12, 2012. We run topic model on ~763K tweets from top 4,029 most frequent users who tweeted about Sandy at least 100 times. We identified 250 well-defined communication patterns based on perplexity. Conversations of most frequent and relevant users indicate the evolution of numerous storm-phase (warning, response, and recovery) specific topics. People were also concerned about storm location and time, media coverage, and activities of political leaders and celebrities. We also present each relevant keyword that contributed to one particular pattern of user concerns. Such keywords would be particularly meaningful in targeted information spreading and effective crisis communication in similar major disasters. Each of these words can also be helpful for efficient hash-tagging to reach target audience as needed via social media. The pattern recognition approach of this study can be used in identifying real time user needs in future crises.

📄 Content

Crisis Communication Patterns in Social Media during Hurricane Sandy

Arif Mohaimin Sadri, Ph.D. Visiting Assistant Professor Civil and Environmental Engineering Rose-Hulman Institute of Technology 5500 Wabash Avenue Terre Haute, IN 47803 Email: sadri.buet@gmail.com
(Corresponding Author)

Samiul Hasan, Ph.D. Assistant Professor Department of Civil, Environmental, and Construction Engineering University of Central Florida 12800 Pegasus Drive, Orlando, FL 32816
Phone: 407-823-2480 Email: samiul.hasan@ucf.edu

Satish V. Ukkusuri, Ph.D. Professor Lyles School of Civil Engineering Purdue University 550 Stadium Mall Drive West Lafayette, IN 47907, USA Phone: (765)-494-2296 Fax: (765)-496-7996 Email: sukkusur@purdue.edu

Manuel Cebrian, Ph.D. Principal Research Scientist Data61, CSIRO 115 Batman Street, West Melbourne VIC 3003, Australia Phone: +61 0403 754 676 Email: Manuel.Cebrian@data61.csiro.au

Word Count: 03 tables + 04 figures + 5697 words = 7447 word equivalents Sadri, Hasan, Ukkusuri, Cebrian

2

ABSTRACT Hurricane Sandy was one of the deadliest and costliest of hurricanes over the past few decades. Many states experienced significant power outage, however many people used social media to communicate while having limited or no access to traditional information sources. In this study, we explored the evolution of various communication patterns using machine learning techniques and determined user concerns that emerged over the course of Hurricane Sandy. The original data included ~52M tweets coming from ~13M users between October 14, 2012 and November 12, 2012. We run topic model on ~763K tweets from top 4,029 most frequent users who tweeted about Sandy at least 100 times. We identified 250 well-defined communication patterns based on perplexity. Conversations of most frequent and relevant users indicate the evolution of numerous storm-phase (warning, response, and recovery) specific topics. People were also concerned about storm location and time, media coverage, and activities of political leaders and celebrities. We also present each relevant keyword that contributed to one particular pattern of user concerns. Such keywords would be particularly meaningful in targeted information spreading and effective crisis communication in similar major disasters. Each of these words can also be helpful for efficient hash-tagging to reach target audience as needed via social media. The pattern recognition approach of this study can be used in identifying real time user needs in future crises.

Keywords: Hurricane Sandy, crisis communication, user concern, topic model, social media, Twitter;
Sadri, Hasan, Ukkusuri, Cebrian

3

BACKGROUND AND MOTIVATION In 2012, New York and New Jersey coastal residents experienced a massive storm surge produced by late season Hurricane Sandy that caused disastrous affects in the mid-Atlantic and northeastern United States across the Atlantic basin (1). Some examples include: $50B in property damage, 72 fatalities, at least 147 direct deaths, destruction of 570K buildings, cancellation of 20K flights, 8.6M power outages in 17 states, 230K cars destroyed due to flooding and thousands of people were displaced from their homes (2-5). Disaster resilience is now a national imperative at all levels with a view to limiting such adverse impacts and more efficient policies are required by the government to allow people to be less dependent on federal resources (6).
As such, vulnerable communities need to respond to any form of disaster with enough preparation (7-10) and effective crisis communication is one major aspect of it. This include systematic planning, collection, organization, and circulation of relevant awareness information to the target audience, reaching out to every individual in a community (11-13). During an emergency, people may obtain weather information from traditional media such as radio or television and social media such as Facebook, Twitter, or the internet. Social media platforms, uniquely different from traditional ones, can help disseminate targeted information during a crisis. However, it is critical to efficiently harness the large-scale and rich information from these communication sources (14). Many studies have acknowledged such value of social media data in emergency response (15-20), community interactions (21; 22), crisis informatics (23-31), and predicting real world actions/events (32; 33).
During Hurricane Sandy, social media also played an important role on crisis communication. New York and New Jersey residents used smart-phones to access social media since they had limited access to traditional information sources (34). Social media communications via social media continued during and after the storm in areas without power (35). Studies have identified authorities, news media, and peers as emergency

This content is AI-processed based on ArXiv data.