New Computational Methods to Study Social Science Questions

Friday, May 11, 2018, 11:00 am - 12:00 pm PDTiCal
6th floor large conference room
This event is open to the public.
AI Seminar
Umashanthi Pavalanathan, Georgia Tech
Video Recording:
Recorded Link
New Computational Methods to Study Social Science Questions
The popularity of online social platforms, including microblogging sites such as Twitter and discussion forums such as Reddit, has generated an explosion of detailed traces of user interaction. These large-scale online traces provide us with an unparalleled opportunity to investigate social science questions at scale, ranging from the spread and adoption of new words to early prediction of disease outbreaks. However, extracting valuable insights from these datasets requires new computational methods to approximate randomized experiments using observational data to infer causal relationships, test existing theories at scale, and develop new theories. In addition, the biases inherent to online social data require careful construction of datasets and appropriate validation of generalizability. In this talk, I will describe novel applications of computational methods to address these challenges in my work. Specifically, I will discuss (1) the biases in using geotagged social data for geo-linguistic inferences, (2) the effects of new linguistic forms in online writing, and (3) the effects of banning online communities promoting negative experiences.
Umashanthi Pavalanathan is a PhD Candidate in the School of Interactive Computing at Georgia Tech. She is a member of the Computational Linguistics Laboratory, working with Dr. Jacob Eisenstein. Her research spans Computational Social Science, Social Computing, and Natural Language Processing. Her thesis work focuses on computational approaches to understanding stylistic variation in online writing. Umashanthi is also interested in applying computation for broader social good and she was a Data Science for Social Good Fellow in summer 2014. Umashanthi did internships at Facebook’s Core Data Science Team, Microsoft Research and the Pacific Northwest National Lab. Before graduate school she was a junior visiting research scholar at the Data to Insight Center at Indiana University, Bloomington. She earned her bachelor’s degree in Engineering specializing in Computer Science and Engineering at University of Moratuwa, Sri Lanka.
« Return to Upcoming Events