Pulling Information from social media in the aftermath of unpredictable disasters

Abstract

Social media have become a primary communication channel among people and are continuously overwhelmed by huge volumes of User Generated Content. This is especially true in the aftermath of unpredictable disasters, when users report facts, descriptions and photos of the unfolding event. This material contains actionable information that can greatly help rescuers to achieve a better response to crises, but its volume and variety render manual processing unfeasible. This paper reports the experience we gained from developing and using a web-enabled system for the online detection and monitoring of unpredictable events such as earthquakes and floods. The system captures selected message streams from Twitter and offers decision support functionalities for acquiring situational awareness from textual content and for quantifying the impact of disasters. The software architecture of the system is described and the approaches adopted for messages filtering, emergency detection and emergency monitoring are discussed. For each module, the results of real-world experiments are reported. The modular design makes the system easy configurable and allowed us to conduct experiments on different crises, including Emilia earthquake in 2012 and Genoa flood in 2014. Finally, some possible functionalities relying on the analysis of multimedia information are introduced.

Technologies Enabling Situational Awareness During Disaster Response: A Systematic Review

“…For example, while a minority (15%) of AARs reported using data analysis technologies during disaster response, newer types of these technologies now exist, such as those using artificial intelligence and machine learning (AI/ML), which have the potential to autonomously ingest, analyze, generate anomaly alerts, and make inferences and conclusions about large volumes of data in real time. Examples included ML analysis of social media posts to detect and localize an incident 59,145,172 and machine vision-based detection of anomalies, such as fire, and prediction about the severity of disaster damage. 175,177 If implemented, AI/ML has the potential to revolutionize SA during disaster response operations.…”

Analysis of Online News Coverage on Earthquakes Through Text Mining

“…More recently, the use of "tweets" (short messages on online social media platform Twitter 1 ) assisted in providing more realtime information on the geographical region of the "news" where the earthquake took place and the amount of structural damage caused (Earle et al, 2010(Earle et al, , 2012Sakaki et al, 2010;Liang et al, 2013;Avvenuti et al, 2015;Bossu et al, 2015;Hicks, 2019). Panagiotou et al (2016) noticed a relationship between the time Twitter users tweeted and the time the event took place, highlighting the importance of social media users acting as news collaborators.…”

“…Liang et al (2013) analyzed how fast the news spread over a period of 90 minutes and managed to identify a correlation between retweet densities and tweeting count per user versus the distance from the earthquake's epicenter. Similarly, Avvenuti et al (2015) studied the relationship between the earthquake's magnitude and how tweets are spread around the geographical area hit by the earthquake. They took into consideration unique Twitter users and the mean value of tweets submitted following the earthquake event.…”

Crisis Mapping During Natural Disasters via Text Analysis of Social Media Messages

“…This limitation drastically reduces the number of useful messages and results in very sparse maps. Recent work have instead demonstrated that emergency reports frequently carry textual references to locations and places [3,4], as shown in Fig. 1.…”

“…http://en.wikipedia.org/wiki/2012 Northern Italy earthquakes 4. http://www.reuters.com/article/2012/05/20/us-quake-italy-idUSBRE84J01K20120520.…”