This talk is jointly sponsored by the Human-Computer Interaction Lab and the Campus Visualization Partnership.

Location:

HCIL (2105 Hornbake, South Wing)

Abstract:

Visualizations can enhance news article content by presenting complex facts clearly and providing contextually-relevant visualizations.  By using novel natural language and text mining approaches, our systems define “queries” that encode the article’s topic (e.g., “unemployment in CA in March,” “global average temperatures in 2012”) and the comparisons that are made in the article’s text (e.g., differences between states or over time) to guide the visualization generation.  Compelling visualizations are relevant and ‘interesting’-concepts that are very hard measure, but we address these challenges in the Contextifier, NewsViews, and PersaLog systems, which are meant to help journalists tell their stories more effectively (joint work with Brent Hecht, Jessica Hullman, Tong Gao, Carolyn Gearig, Josh Ford, and Nick Diakopoulos).

Bio:

Eytan Adar is an Associate Professor in the School of Information & Computer Science and Engineering at the University of Michigan.   He works at the intersection of HCI and IR/Data Mining and ranges from empirical studies of large-scale online behaviors to building new systems, tools and methods.  He has a Bachelors and Masters from MIT and a PHD in Computer Science at the University of Washington. He was a researcher at HP Labs and Xerox PARC, and spun out a company called Outride.   Eytan is co-founder of ICWSM and has served as general chair for ICWSM and WSDM.  His website is  http://www.cond.org