BBL: Combining Crowds and Computation to Make Discoveries and Solve Mysteries
Place: HCIL (room 2105 Hornbake, South Wing)
We are living in the era of big data, and making sense of this data to improve the human condition is a major challenge. Automated techniques in machine learning, natural language processing, computer vision, and other areas have made significant headway, but many types of complex data analysis still require human intervention. Crowdsourcing and human computation raise exciting possibilities for enhancing computational data analysis techniques with scalable human intelligence and creativity, allowing us to solve harder problems and generate deeper insights than humans or computers working alone. In this talk, I will describe several of my recent projects exploring the potential of crowdsourced data analysis. These include Crowdlines, a system that crowdsources a comprehensive overview of a knowledge domain using existing material gathered from the web; Incite, a system that engages non-expert crowds in helping professional scholars make discoveries in large collections of historical documents; and Context Slices, a system that combines crowdsourcing and visual analytics techniques to help experts solve mysteries, such as identifying the subject matter in historical photos or uncovering a terrorist plot in a body of textual evidence.
Kurt Luther is Assistant Professor of Computer Science at Virginia Tech, where he is also Co-Director of Social Informatics for the Center for Human-Computer Interaction. He builds and studies social technologies that support creativity and discovery, often with applications to the creation and analysis of visual media, such as animation, graphic design, and photography. He also explores how social technologies can engage the public in historical research, preservation, and education. His work is currently funded by the National Science Foundation, the National Archives, and Google. Previously, he was a postdoc in the HCI Institute at Carnegie Mellon University, and he holds a Ph.D. in Human-Centered Computing from Georgia Tech.