home · mobile · calendar · bactac · 2005-2006 · 

BACTAC - White

Automatic Identification of Conflicting Statements in Natural Language Text
PhD Candidate

In biology, as in many other fields, the steadily rising volume of publications has made it increasingly difficult for scientists to keep up with the relevant literature. This has inspired many techniques for automatically extracting information from biological publications. However, little has been done to address the problem of conflicting research reports. It is quite common for one paper to refute another directly, or for several papers to disagree, directly or indirectly, with each other. Often, this disagreement is explicitly noted by the authors. I propose a system that will seek out mentions of conflict in scientific papers, trace the conflicting references back to their sources, and then summarize and present them to the user in graphical form for further analysis.

Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
May 5, 2012 (14:24)