Mathematical and Computational Models for Crowdsourced Geolocation

Social media generate large amounts of almost real-time data which can turn out extremely valuable in an emergency situation, specially for providing information within the first 72 hours after a disaster event. Despite there is abundant state-of-the-art machine learning techniques to automatically classify social media images and some work for geolocating them, the operational problem in the event of a new disaster remains unsolved. Currently the state-of-the-art approach for dealing with these first response mapping is first filtering and then submitting the images to be geolocated to a crowd of volunteers, assigning the images randomly to the volunteers. In this work we evaluate the potential of artificial intelligence (AI) to assist emergency responders and disaster relief organizations in geolocating social media images from a zone recently hit by a disaster by (i) building a simple model of the capabilities of each of the volunteers of the crowd; and (ii) intelligently assigning tasks to those volunteers which can perform them better. We present in this paper some new methods that outperform random allocation of tasks in a simplified, scalable scenario. Moreover, we show that for a given set of tasks and volunteers, we are able to process them with a significantly lower annotation budget, that is, we are able to make less volunteer solicitations without loosing any quality on the final consensus.