CIME: Context-aware geolocation of emergency-related posts

Abstract

Information extracted from social media has proven to be very useful in the domain of emergency management. An important task in emergency management is rapid crisis mapping, which aims to produce timely and reliable maps of affected areas. During an emergency, the volume of emergency-related posts is typically large, but only a small fraction is relevant and help rapid mapping effectively. Furthermore, posts are not useful for mapping purposes unless they are correctly geolocated and, on average, less than 2% of posts are natively georeferenced. This paper presents an algorithm, called CIME, that aims to identify and geolocate emergency-related posts that are relevant for mapping purposes. While native geocoordinates are most often missing, many posts contain geographical references in their metadata, such as texts or links that can be used by CIME to filter and geolocate information. In addition, social media creates a social network and each post can be enhanced with indirect information from the post’s network of relationships with other posts (for example, a retweet can be associated with other geographical references which are useful to geolocate the original tweet). To exploit all this information, CIME uses the concept of context, defined as the information characterizing a post both directly (the post’s metadata) and indirectly (the post’s network of relationships). The algorithm was evaluated on a recent major emergency event demonstrating better performance with respect to the state of the art in terms of total number of geolocated posts, geolocation accuracy and relevance for rapid mapping.

TriggerCit: Early Flood Alerting using Twitter and Geolocation -- a comparison with alternative sources

“…Like in other papers analysing social media, the post geolocation is extracted from Twitter metadata. As studied in the literature (e.g., Scalia et al 2021), this is usually available only in a small percentage of the posts. Recent papers focused on analysing in parallel several sources of information for deriving awareness on emergency events.…”

“…A more comprehensive approach to social media posts geolocation has been proposed in the CIME system (Scalia et al 2021). The NER is based on Polyglot (Al-Rfou et al 2015), which supports more than 40 languages.…”

“…The CIME geolocation algorithm described in Scalia et al 2021 is then applied to associate tweets with geographical entities, whenever possible. Locations in unrelated countries are filtered out using national administrative boundaries.…”

Analyzing social media with crowdsourcing in Crowd4SDG

“…Relevance can depend on the use of the information and the actors involved; therefore, its definition is part of a data analysis project description, as it will be discussed in the following of this paper. The extraction process requires several phases of data preparation and analysis, as raw datasets extracted from social media usually contain a very low percentage of useful information, estimated to be between 0.5% and 3% in the literature [30,41]. However, even a very small number of relevant posts can be useful in providing a significant overview on the areas in which the event occurs [14,43].…”

“…The main goal is to improve the overall quality of the dataset. VisualCit also provides automatic geolocation functionalities, based on OpenStreeMap [20] and the CIME approach [41] for geocoding. The CS Project Builder provides an environment for crowdsourcing based on the open-source PyBossa framework.…”