Problems involving social networks and massive social datasets motivate some of the most difficult computational challenges today, and there is a pressing demand for understanding the structure of these problems. The design and measurement of social networks is commonly studied for properties shared with other networks, when in fact many aspects are specifically characteristics of social contexts. The spread of ideas is often studied as an epidemiological process, but very little research has examined how individuals make social decisions when adopting ideas. In the online world, experimentation using A/B tests does not account for social interference between treatment units, making it poorly suited for many studies on social networks. In this talk I will present research on these challenges (social structure, social contagion, social experimentation), developing computational approaches adapted to understanding the structure of social networks and social processes, and also to impacting the design of large-scale online social systems.
Johan Ugander is a Ph.D candidate at Cornell University, advised by Jon Kleinberg. His research develops tools for analyzing massive social graphs and other large-scale social data, aiming to provide a better understanding of social structure and human decision making while also impacting the design of large-scale computational systems. He is the recipient of the WSDM '13 Best Student Paper Award and the WebSci '12 Best Paper Award, and his work has been featured in popular-media outlets including The New York Times, The Economist, NPR, and Wired.