Abstract:
This talk emphasizes the use of structured approaches towards addressing
Natural Language Understanding (NLU) problems. We argue that many NLU tasks
can benefit from using models that are capable of incorporating not just
linguistic cues, but also the contexts in which these cues appear. In this
talk, we present a structured approach to model the 'flow of information'
in text to solve two seemingly distinct problems: (i) Identifying need for
instructor intervention in MOOC discussion forums, and (ii) Analyzing a
paragraph to identify if a desire expressed in it was fulfilled.

In the first problem, we analyze contents of online educational discussion
forums to automatically suggest threads to instructors that require their
intervention. This can alleviate the need for the instructor to manually
peruse all threads of the forum, and help students who need to interact
with the instructor. Our method incorporates thread structure for the
problem by using latent variables that abstract contents of individual
posts and model the flow of information in the thread.

We then consider the problem of reading and understanding a textual
paragraph containing an expression of a desire to identify if the desire
was fulfilled. The method interprets the paragraph as a story from the
perspective of the protagonist - the entity that expressed the desire. We
track the protagonist's actions and emotional states to make the binary
prediction.


Speaker Bio:
Snigdha Chaturvedi is a PhD candidate in Computer Science at University of
Maryland, College Park working with Dr. Hal Daume III. She is supported by
the IBM PhD Fellowship and NSF grants. Her primary research interests are
in solving problems related to Natural Language Understanding. Before
joining UMD, she was a Blue Scholar at IBM Research Labs in Delhi. She
earned her undergraduate degree from IIT Kanpur in 2009. Her BTech Project
(guided by Dr. Harish Karnick) was on predicting cardiological conditions
from automatic analysis of ECG signals, and won the Binay Kumar Sinha award.