Talk by Darren Gergle, 14th Sept, 13:15, BC 410

Every day people produce an extraordinary amount of user-generated content in social media systems such as Wikipedia, Flickr, and Facebook. There is great opportunity for such data to help provide global coverage of a variety of topics. Yet, there is also a risk that language barriers and coverage biases present in the data can limit the information available to certain user communities. In this talk, I will present research that illustrates both the challenges and opportunities of user-generated content in the context of multilingual Wikipedia. First, I will present studies that empirically assess the content diversity and coverage biases that exist across numerous language editions of Wikipedia. Next, I will demonstrate the effects these coverage biases can have on end-user technologies. Finally, I will introduce Omnipedia—a system our lab developed that highlights the opportunities such data present for designing a new class of global technologies.

Darren Gergle (Ph.D., Carnegie Mellon University) is a Professor in Communication Studies and Electrical Engineering and Computer Science (by courtesy) at Northwestern University. His research focuses on social computing and human-computer interaction and applies cognitive and social psychological theories to the design, development and evaluation of novel communication technologies. His research is generously supported by grants from the National Science Foundation, National Institute of Mental Health, Google, Microsoft Research and Facebook.