Benjamin Herfort

A geographic approach for combining social media and authoritative data towards identifying useful information for disaster management

In recent years, social media emerged as a potential resource to improve the management of crisis situations such as disasters triggered by natural hazards. Although there is a growing research body concerned with the analysis of the usage of social media during disasters, most previous work has concentrated on using social media as a stand-alone information source, whereas its combination with other information sources holds a still underexplored potential. This article presents an approach to enhance the identification of relevant messages from social media that relies upon the relations between georeferenced social media messages as Volunteered Geographic Information and geographic features of flood phenomena as derived from authoritative data (sensor data, hydrological data and digital elevation models). We apply this approach to examine the micro-blogging text messages of the Twitter platform (tweets) produced during the River Elbe Flood of June 2013 in Germany. This is performed by means of a statistical analysis aimed at identifying general spatial patterns in the occurrence of flood-related tweets that may be associated with proximity to and severity of flood events. The results show that messages near (up to 10 km) to severely flooded areas have a much higher probability of being related to floods. In this manner, we conclude that the geographic approach proposed here provides a reliable quantitative indicator of the usefulness of messages from social media by leveraging the existing knowledge about natural hazards such as floods, thus being valuable for disaster management in both crisis response and preventive monitoring.

Exploring the Geographical Relations Between Social Media and Flood Phenomena to Improve Situational Awareness

Recent research has shown that social media platforms like twitter can provide relevant information to improve situation awareness during emergencies. Previous work is mostly concentrated on the classification and analysis of tweets utilizing crowdsourcing or machine learning techniques. However, managing the high volume and velocity of social media messages still remains challenging. In order to enhance information extraction from social media, this chapter presents a new approach that relies upon the geographical relations between twitter data and floodphenomena.Ourapproachusesspecificgeographicalfeatureslikehydrological data and digital elevation models to prioritize crisis-relevant twitter messages. We apply this approach to examine the River Elbe Flood in Germany in June 2013. The results show that our approach based on geographical relations can enhance informationextractionfromvolunteeredgeographicinformation,thusbeingvaluable for both crisis response and preventive flood monitoring.

The Tasks of the Crowd: A Typology of Tasks in Geographic Information Crowdsourcing and a Case Study in Humanitarian Mapping

In the past few years, volunteers have produced geographic information of different kinds, using a variety of different crowdsourcing platforms, within a broad range of contexts. However, there is still a lack of clarity about the specific types of tasks that volunteers can perform for deriving geographic information from remotely sensed imagery, and how the quality of the produced information can be assessed for particular task types. To fill this gap, we analyse the existing literature and propose a typology of tasks in geographic information crowdsourcing, which distinguishes between classification, digitisation and conflation tasks. We then present a case study related to the “Missing Maps” project aimed at crowdsourced classification to support humanitarian aid. We use our typology to distinguish between the different types of crowdsourced tasks in the project and choose classification tasks related to identifying roads and settlements for an evaluation of the crowdsourced classification. This evaluation shows that the volunteers achieved a satisfactory overall performance (accuracy: 89%; sensitivity: 73%; and precision: 89%). We also analyse different factors that could influence the performance, concluding that volunteers were more likely to incorrectly classify tasks with small objects. Furthermore, agreement among volunteers was shown to be a very good predictor of the reliability of crowdsourced classification: tasks with the highest agreement level were 41 times more probable to be correctly classified by volunteers. The results thus show that the crowdsourced classification of remotely sensed imagery is able to generate geographic information about human settlements with a high level of quality. This study also makes clear the different sophistication levels of tasks that can be performed by volunteers and reveals some factors that may have an impact on their performance.