Abstract: Recent years have witnessed multiple international protest movements which have purportedly been greatly affected by the use of Twitter, a micro-blogging platform. Social movement actors in Iran, Moldova, Kyrgyzstan and Thailand are thought to have utilized Twitter to spread information, co-ordinate protest activities, evade government censorship and, in some cases, to spread misinformation. We propose a framework for conceptualizing and analyzing Twitter data related to contentious collective action crises. Our primary research goal is to delineate a framework informed with a social movements lens and to demonstrate the framework by means of Twitter usage data related to the Thailand protests of 2010. Our proposed framework concerns itself with two aspects of protest activities and Twitter usage, namely, analyzing the content and structure of messages and our construct of Twitter protest waves.

Abstract: This paper offers a descriptive account of Twitter (a micro-blogging service) across four high profile, mass convergence events-two emergency and two national security. We statistically examine how Twitter is being used surrounding these events, and compare and contrast how that behavior is different from more general Twitter use. Our findings suggest that Twitter messages sent during these types of events contain more displays of information broadcasting and brokerage, and we observe that general Twitter use seems to have evolved over time to offer more of an information-sharing purpose. We also provide preliminary evidence that Twitter users who join during and in apparent relation to a mass convergence or emergency event are more likely to become long-term adopters of the technology.

Abstract: Informal communication channels are often the primary means by which time-sensitive hazard information first reaches members of the public. The capacity for informal communications has been recently transformed by the widespread adoption of social media technologies, such as the micro-blogging service Twitter, which allows individuals to interact with a broad audience over great distances. During a disaster or crisis event, this networked communication mechanism provides a means to communicate information and facilitate collaboration both locally and among distributed networks. This paper examines the use of Twitter following a technological disaster, showing how geographically dispersed individuals broadcast information about the impact of the disaster and its long-term effects, in contrast with the dearth of participation among public officials and industry representatives. Non-local users challenged authoritative accounts of the disaster and corrected misinformation. Conclusions are provided for policy makers and suggestions are offered for further research.

Abstract: Microblogging services like Twitter afford opportunities for real time determination of situation awareness during crises as people report, via their statuses, information about events on the ground. An important component of the information included in a tweet are mentions of place names that may be sites of damage, injuries, or relief efforts. Methods for extracting these place names and determining the actual location being referenced are an essential part of the suite of tools required for automated extraction of situation awareness from tweets. Extracting and disambiguating place name mentions from text have been areas of extensive research. Twitter, however, presents challenges given the 140 character restriction on status and the informal, abbreviated language that are a norm in this communication channel. Named entity recognizers, which are dependent on labeled training data, may not be useful in this medium for extracting location mentions because the typical training domains for these taggers are absent the noise found in Twitter statuses. Additionally, the contextual information that is necessary for disambiguating place names is not always present. In this paper, we demonstrate a new technique, RapidGeo, for extracting and disambiguating place names from a location specific Twitter feed using an unsupervised technique for tagging location mentions and relying on the known geographic context of the feed for disambiguation. Our location tagging technique performs much better than an off-the-shelf named entity recognizer and we achieve reasonable precision in disambiguating extracted place names. We argue that such fast, high precision, unsupervised approaches are needed when important, actionable information is required from noisy data sources such as Twitter. © 2012 ISCRAM.

Abstract: This paper focuses on the micro-blogging service Twitter, looking at source credibility for information shared in relation to the Fukushima Daiichi nuclear power plant disaster in Japan. We look at the sources, credibility, and between-language differences in information shared in the month following the disaster. Messages were categorized by user, location, language, type, and credibility of information source. Tweets with reference to third-party information made up the bulk of messages sent, and it was also found that a majority of those sources were highly credible, including established institutions, traditional media outlets, and highly credible individuals. In general, profile anonymity proved to be correlated with a higher propensity to share information from low credibility sources. However, Japanese-language tweeters, while more likely to have anonymous profiles, referenced low-credibility sources less often than non-Japanese tweeters, suggesting proximity to the disaster mediating the degree of credibility of shared content. © 2012 ISCRAM.

Abstract: We present Tweet4act, a system to detect and classify crisis-related messages communicated over a microblogging platform. Our system relies on extracting content features from each message. These features and the use of an incident-specific dictionary allow us to determine the period type of an incident that each message belongs to. The period types are: Pre-incident (messages talking about prevention, mitigation, and preparedness), during-incident (messages sent while the incident is taking place), and post-incident (messages related to the response, recovery, and reconstruction). We show that our detection method can effectively identify incident-related messages with high precision and recall, and that our incident-period classification method outperforms standard machine learning classification methods.

Abstract: Microblogging sites such as Twitter can play a vital role in spreading information during “natural” or man-made disasters. But the volume and velocity of tweets posted during crises today tend to be extremely high, making it hard for disaster-affected communities and professional emergency responders to process the information in a timely manner. Furthermore, posts tend to vary highly in terms of their subjects and usefulness; from messages that are entirely off-topic or personal in nature, to messages containing critical information that augments situational awareness. Finding actionable information can accelerate disaster response and alleviate both property and human losses. In this paper, we describe automatic methods for extracting information from microblog posts. Specifically, we focus on extracting valuable “information nuggets”, brief, self-contained information items relevant to disaster response. Our methods leverage machine learning methods for classifying posts and information extraction. Our results, validated over one large disaster-related dataset, reveal that a careful design can yield an effective system, paving the way for more sophisticated data analysis and visualization systems.

Abstract: Social media allows the public to engage in the disaster response and recovery process in new and exciting ways. Many emergency management agencies in the United States are embracing social media as a new channel for alerts, warnings, and public outreach, but very few are mining the massive amounts of data available for use in disaster response. The research reflected in this paper strives to help emergency management practitioners harness the power of community volunteered information in a way that is still novel in most parts of the country. Field verification and research combined with survey results attempts to identify and solve many of the barriers to adoption that currently exist. By helping practitioners understand the virtues and limitations of this type of data and information, this research will encourage the use of community volunteered information in the emergency operations center.

Abstract: Real-time information from microposts like Twitter is useful for applications in the crisis management domain. Currently, that potentially valuable information remains mostly unused by the command staff, mainly because the sheer amount of information cannot be handled efficiently. Sentiment analysis has been shown as an effective tool to detect microposts (such as tweets) that contribute to situational awareness. However, current approaches only focus on two or three emotion classes. But using only tweets with negative emotions for crisis management is not always sufficient. The amount of remaining information is still not manageable or most of the tweets are irrelevant. Thus, a more fine-grained differentiation is needed to identify relevant microposts. In this paper, we show the systematic evaluation of an approach for sentiment analysis on microposts that allows detecting seven emotion classes. A preliminary evaluation of our approach in a crisis related scenario demonstrates the applicability and usefulness.

Abstract: The Four Phase Model of Emergency Management has been widely used in developing emergency/disaster response plans. However, the model has received criticism contrasting the clear phase distinctions in the model with the complex and overlapping nature of phases indicated by empirical evidence. To investigate how phases actually occur, we designed PhaseVis based on visualization principles, and applied it to Hurricane Isaac tweet data. We trained three classification algorithms using the four phases as categories. The 10-fold cross-validation showed that Multi-class SVM performed the best in Precision (0.8) and Naïve Bayes Multinomial performed the best in F-1 score (0.782). The tweet volume in each category was visualized as a ThemeRiver[TM], which shows the 'What' aspect. Other aspects – 'When', 'Where', and 'Who' – Are also integrated. The classification evaluation and a sample use case indicate that PhaseVis has potential utility in disasters, aiding those investigating a large disaster tweet dataset.

Abstract: In this paper we present a new approach to enhance information extraction from social media that relies upon the geographical relations between twitter data and flood phenomena. We use specific geographical features like hydrological data and digital elevation models to analyze the spatiotemporal distribution of georeferenced twitter messages. This approach is applied to examine the River Elbe Flood in Germany in June 2013. Although recent research has shown that social media platforms like Twitter can be complementary information sources for achieving situation awareness, previous work is mostly concentrated on the classification and analysis of tweets without resorting to existing data related to the disaster, e.g. catchment borders or sensor data about river levels. Our results show that our approach based on geographical relations can help to manage the high volume and velocity of social media messages and thus can be valuable for both crisis response and preventive flood monitoring.

Abstract: On the morning of April 15, 2013, an Improvised Explosive Device (IED) was detonated near the finish line of the Boston Marathon, resulting in a large number of casualties. This generated a week-long response under the US National Incident Management System. In this paper, we examine online, terse messages broadcast by responding organizations and their amplification by other official entities via retransmission. Content analysis of official messages shows strong similarities with posting patterns previously observed in response to natural hazards, with the primary exception of themes related to the criminal investigation, suggesting a possible revision of guidelines for public information in light of the needs arising from extended counterterrorism operations undertaken in an urban environment. Network analysis demonstrates message posting and amplification were dominated by local actors, underscoring the importance of local readiness for management of official public information activities in the context of extremely high-profile events.

Abstract: This research presents a centralized boundary object website and mobile app focused on allowing participants to participate in developing an early warning system through space weather and the beauty of the aurora borealis. Because of the beauty and majesty of auroral activity, people will seek information about when and where these unpredictable events occur. This activity, commonly referred to as nowcasting, can be combined with scientific data collected from observatories and satellites and serve as an early warning system with potentially far greater accuracy and timeliness than the current state of the art. We believe that long-term engagement with a citizen science tool will help bridge the many social worlds surrounding the aurora borealis and lead to the development of an early warning system that may correlate the visibility of the northern lights to violent space weather. We hope this will lead to other real time crowdsourced early warning systems in the future.