Abstract
The most recent catastrophic events, from the 2010 Haiti earthquake to the devastating 2013 Colorado floods, have shown a strong adoption of social media platforms by ordinary people. The data and meta-data produced by the users during and after the extra-ordinary situations could have enormous potentialities if integrated with the traditional systems for emergency management and used for hyperlocal situational awareness. The great majority of the current literature is focused on Twitter for several reasons strictly linked to the architectures and practices of use of the platform itself. It is possible to classify the existing systems based on the analysis of Twitter data at least in three different categories: 1) semantic systems; 2) meta-data systems; and 3) smart self-learning systems. In this contribution, a review of the most significant and important tools used to analyze Twitter data will be presented and an innovative and smart solution will be proposed for future development.
TopBackground
Social media platforms are built from the beginning to be used socially, and oriented around collaboration and sharing. These potentialities are emphasized in extra-ordinary contexts, when ordinary people adopt these tools to provide or search for first-hand and real-time information regarding a certain event (i.e. an earthquake, flood, etc.) (Lindsay, 2011; Taylor et al., 2012). The most recent catastrophic events, from the 2010 Haiti earthquake to the devastating 2013 Colorado floods, in fact, have shown that these platforms have been strongly used both during and after disasters (Figure 1), allowing a real-time dissemination of information to the wider public, an effective situational awareness, and an up-to-date picture of what is happening on the ground (i.e. Farinosi & Micalizzi, 2013; White et al., 2014).
According to Kotsiopoulos (2014), in extra-ordinary situations, social media enable citizens to play at least three roles: 1) first responders/volunteers; 2) citizen journalists/reporters; and 3) social activists. Oftentimes citizens on the scene experience the event first-hand and are able to provide updates more quickly than disaster response organizations and traditional news media (Sweetser & Metzgar, 2007; Procopio & Procopio, 2007; Farinosi & Micalizzi, 2013).
Given the increasing availability of data and meta-data produced and/or distributed on these online platforms, it is pivotal to understand how they should be used and integrated with traditional systems for situational awareness, supporting in this way the work of Civil Protection, Red Cross, Fire Department, and other agencies.
Key Terms in this Chapter
Tweet: Message posted on Twitter which may contain up to 140 characters of text, photos, videos, links, and so on.
Meta-Data: Data providing information about a certain item’s content. For example, a tweet may include meta-data specifying GPS coordinates, or information about when it was sent, by who, and so on, which can be analyzed to extract meaningful information.
Data Mining: A computational process of discovering patterns in large data sets. The main goal is to extract knowledge from a data set and transform it into an understandable structure for further use (i.e. machine learning or predictive analytics).
Twitter APIs: A set of software libraries used by developers to extract data from Twitter streams. Twitter provides different types of software libraries (e.g. REST APIs, Streaming APIs, etc.), each one with its own specific function.
Web Crawler: Software program able to automatically analyze the content of web pages in order to recognize useful information.
Hashtag: Annotation format represented by the “#” symbol, used to indicate with a single word (or a combination of words) the core meaning of a tweet.
Situation Awareness: Pre-requisite state of knowledge and cognition of events useful for making decisions in situations involving uncertainty and crisis.