Combining Topic Structure with Rhetorical Information for Automatic Summarization

In this talk, I will give an overview of automatic summarization and present some of my current research, which combines topic structure with rhetorical information to create improved summaries. Summarization involves finding the most important information in a text in order to convey the meaning of the document. In my work, I propose a method for using topic information to influence which content is selected for a summary.

Natalie Schrimpf  graduated from Dartmouth in 2012 with a major in linguistics and minors in French and Psychology. My senior thesis, "An Acoustic Sociophonetic Analysis of Middle Tennessee English Dialect Features Across Different Social Groups," looked at how Southern speech correlates with social characteristics such as education level and political views. Since graduating, I have been a PhD student in the linguistics and natural language processing. My dissertation focuses on automatic text summarization.

This event is free and open to the public.