Kurt Luther, Professor of Computer Science and History is visiting Segal to meet students and discuss his work with expert led crowdsourcing.

Solving Photo Mysteries with Expert-Led Crowdsourcing

How can expert-led crowdsourcing, an approach combining the strengths of individual experts and crowds of novices, be used to solve photo mysteries? We conducted a qualitative study with expert investigators in journalism, national security, and human rights organizations to understand how they perform image geolocation. This process of hunting down the precise location on Earth where a photo or video was created is complex, yet crucial for verifying or debunking claims about visual evidence. Informed by this study, we built GroundTruth, a system where experts collaborate with crowds to geolocate unknown images. Our experiments found that by adapting an expert diagramming technique, crowds can narrow down a search area by half, with near-perfect accuracy, in 10 minutes.

Background

Kurt Luther is an assistant professor of computer science and, by courtesy, history at Virginia Tech. He directs the Crowd Intelligence Lab an interdisciplinary research group exploring how crowdsourcing systems can support creativity and discovery. His research is supported by the NSF (including a CAREER Award), the NIH, the NEH, the National Archives, and Google.