I am a PhD student in Computer Science at the University of Utah. I am originally from Kentucky, where I graduated with a degree in Computer Science from Morehead State University. I enjoy living in Utah a great deal and am also an Artificial Intelligence enthusiast and electric bassist.
My advisor is Ellen Riloff and I am a member of the Natural Language Processing research group, where my current research focus is on automatic coreference resolution. Specifically, I am interested in integrating real-world knowledge with a learning-based approach to this problem. Other problem areas that I am interested in are information extraction, document summarization and unsupervised machine learning.
Often in Natural Language Processing tasks we are presented with the problem of handling phrases or words that may refer to another phrase that has previously occurred in a text. These "referring" or "pointing-back" phrases are called anaphora and what they refer to are called antecedents. If an anaphor and its antecedent refer to the same entity in the "real-world" then they are considered coreferential. For an example consider the following sentence:
David Beckham is the LA Galaxy midfielder.
The phrase the LA Galaxy's midfielder is an example of an anaphor with its antecedent being David Beckham, since both refer to the same entity in the ``real-world'' they are coreferential. The act of figuring out which anaphora corefer to which antecedents is called coreference resolution.
My current work includes further developing the Reconcile system, which is a joint effort between Cornell, Utah and Lawrence-Livermore National Labs to create an extensible system for identifying coreferent textual entities. More information about Reconcile will be available in the coming months. My long-term objectives include exploring the use of knowledge-rich features (including the web) towards improving unsupervised coreference resolution. I'm always looking for new applications of coreference resolution too...
For the fall semester I am enrolled in the following course.
MATH 5750 - Optimization - MWF (9:40 - 10:30) with Dr. Vasquiez
On Monday through Friday you can usually find me in the Natural Language Processing Lab. If you're interested in such things, stop by the Machine Learning Reading Group on Wednesdays at 3:40 in the Graphics Annex conference room. Also check out the Algorithms Seminar, I hang out in there too. I have also organized the NLP seminar in the past, check back in the fall for an updated schedule.
Nathan Gilbert
School of Computing, University of Utah
50 S. Central Campus Dr., Room 3190
Salt Lake City, UT 84112
Phone: (801) 656-7268
