Crowdsourcing Supervised Classification for Coastline Flooding through the Use of Flickr Contributions
Date
2014-10-08
Authors
Wells, Carter L
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis describes the role that social media can play in improving the classification of remotely sensed images of disaster areas, as well as presenting a new approach for using crowdsourced information to help assess flooding resulting from natural disasters. Hurricane Sandy is used as a case study, and the social media platform that we chose to use was Flickr, an open-source image content site. Remotely sensed images were used to determine the area of flooding that occurred in affected parts of New York and New Jersey, and then the Flickr images were examined to determine where observers were posting from. In the process of researching and writing this thesis, we conducted a literature search of the use of open-source social media platforms as non-authoritative sources of information in times of disaster, and the role they play in the response to disasters. This thesis hopes to expand on previous research on the academic uses for social media, as well as add to the expanding role that social media can play in times of disaster by presenting a new approach to classifying images using crowdsourced data.
Description
Keywords
Social media, Flickr, Crowdsourcing, Flood mapping, Coastal flooding