College of Liberal Arts

Katrin Erk's Research Paves Way for Language Technology Applications

Wed, Jul 29, 2015
Katrin Erk joined the Department of Linguistics in 2006.
Katrin Erk joined the Department of Linguistics in 2006.

Katrin Erk, an associate professor in the Linguistics Department, has been awarded a National Science Foundation (NSF) grant for her project titled "Deep Natural Language Understanding with Probabilistic Logic and Distributional Similarity.” The three-year award of $408,287 begins September 2015 and comes from NSF’s Directorate for Computer & Information Science & Engineering.

Here’s the problem that Erk will address in her research: The Web offers huge amounts of information, but this makes it hard to find and extract relevant information. Natural language processing has made huge strides in developing tools that extract information and automatically answer questions.

Contemporary intelligent systems have long used logic to describe precisely what a sentence means and how its pieces connect. But this precision has a downside: Logic needs the data to exactly match its expectations, or it breaks down. This is problematic for applications like question answering because language is hugely variable. There are often many different ways to say the same thing, or to say things that are not exactly the same but similar enough to be relevant.

To address this problem, Erk's project combines logic with a technology that identifies words and passages that are similar, but that are not exact matches. Furthermore, language often only implies things rather than stating them outright. The project handles this through a mechanism that draws conclusions that are likely but not 100 percent certain, and that states its level of confidence in a conclusion.

In her research, Erk says she hopes to forge new links between computational and theoretical linguistics by transferring ideas in both directions. Through its combination of precision and approximation, this project paves the way for language technology applications that understand language more deeply and thus will be better able to extract information and automatically answer questions.

 

 

Bookmark and Share