skip to main content

Virginia Tech News / Articles / 2014 / 12 

Army awards Discovery Analytics Center researcher $300,000 to develop Big Data storytelling tool

December 17, 2014

Chang-Tien Lu, an associate professor of computer science in the College of Engineering at Virginia Tech, National Capital Region, has been awarded a $300,000 subcontract from the United States Army Research Office and United States Army Engineer Research and Development Center to develop an automated tool to make sense of data captured in news articles, tweets, images, and audio and video streams. 

Lu, with co-principal investigators Ing-Ray Chen, a professor in the Department of Computer Science, and Naren Ramakrishnan, the Thomas L. Phillips Professor and director of the Discovery Analytics Center of the Department of Computer Science, will oversee research that may one day help law enforcement agencies connect suspects and events.

“The goal of this project is to become the gold standard by which the intelligence community measures its analytical decisions and quantitative predictions,” Lu said. “By developing efficient spatiotemporal storytelling techniques we will discover meaningful relationships among entities and deliver rapid and accurate insights.”

By extracting information about location, identity, organization, and relationship, Lu will provide a way to rapidly generate “linkages.”

For example, the location of an event can be extracted and coded from a document, text, or tweet. Codes will decipher the time stamp of an event, connect names, and link entities.

The data sets Lu is interested in are huge, which makes finding linkages even more difficult. Lu will tackle this problem by using open-source software that can process data across several computer clusters, as well as a method that increases computing power by employing graphics processing units.

In that manner, the same graphics cards that produce slick visual effects in modern video games handle the computations required to analyze huge data sets.

The system will provide analysts with a variety of visual representation schemes for the graphs generated by the system.

Lu is the associate director of the Discovery Analytics Center in Arlington, Virginia. It is an interdisciplinary hub with operations on both the Blacksburg campus and at the Virginia Tech National Capital Region in Arlington. 

The center receives support from the Institute for Critical Technology and Applied Science at Virginia Tech.

 

 

Written by Emily Kathleen Alberts

Contact: