Button to scroll to the top of the page.

Updates

Campus health and safety are our top priorities. Get the latest from UT on COVID-19.

Get help with Zoom and more.

Events Calendar

Yearly View
View as List
Monthly View
By Month
Weekly View
By Week
Daily View
Today
Search
Search
Download as iCal file
SDS Seminar Series: Vera Liu (Friday 4/9/21, 2pm)
Friday, April 09, 2021, 02:00pm - 03:00pm

The Spring 2021 SDS Seminar Series continues on Friday, April 9 from 2:00 p.m. to 3:00 p.m. via Zoom with Vera Liu (PhD student at the Department of Statistics and Data Sciences in the College of Natural Sciences at The University of Texas at Austin).

Please contact stat.admin@austin.utexas.edu for the Zoom link.

SDS Logo
TitleAssisted structural topic modeling with multiple sources

Abstract: In this project, our goal is to make meaningful inferences from multiple datasets of different nature. We try to understand if different election candidates focus on different topics when they speak (in the form of tweets) to voters prior to the election. Our data includes campaign tweets from UK general election candidates, newspaper quotes from The Guardian and The Daily Telegraph with annotated topics, and UK parliament bills.  With the assumption that candidates focus on different issues based on their covariate levels (for example, candidate’s gender, party affiliation and the district they are from), we build a joint Bayesian model that extends the recently developed structural topic model (Roberts et al, 2016). The joint model enables us to borrow information from sources such as newspaper quotes, which helps to identify current political issues of interest. We try to answer questions such as: Do incumbent candidates focus on different issues than candidates who are not? Do candidates coming from a region with lower quality of healthcare put more emphasis on healthcare issues compared to other topics?