CIVICA Applied Data Science Seminar Series

Session 4 - Spring 2022

Session 4 - Spring 2022 will take place on Wednesday, 4:00 PM CET 23 February, on Zoom Meeting

About The Series

The CIVICA Data Science Seminar series is a unique multi-disciplinary series focused on applications and methodologies of data science for the social, political, and economic world.

Session 4 Spring 2022 Details

"Political scientists commonly seek to make statements about how a word’s use and meaning varies over circumstances—whether that be time, partisan identity, or some other document-level covariate. A promising avenue is the use of domain-specific word embeddings, that simultaneously allow for statements of uncertainty and statistical inference."

In this session of the CIVICA Data Science Seminar Series, Prof. Arthur Spirling, Professor of Politics and Data Science at New York University will introduce the a la Carte on Text (ConText) embedding regression model for this purpose and evaluate how this method outperforms well-known competitors for studying changes in meaning of words across groups and time.

Seminar Speaker

Prof. Arthur Spirling
Prof. Arthur Spirling

Arthur Spirling is the Professor of Politics and Data Science at New York University. He also acts as the Deputy Director and the Director of Graduate Studies (MSDS) at the Center for Data Science. He specialises in political methodology and legislative behavior, with an interest in the application of text-as-data/NLP, Bayesian statistics, machine learning, item response theory and generalized linear models in political science.


Agenda

Dr. Olga Gasparyan

Welcome Introduction Dr. Olga Gasparyan, Hertie School

Setting the scene: Brief intro to the speaker and his talk

Prof. Arthur Spirling

Seminar Session Prof. Arthur Spirling, New York University

Embedding Regression: Models for Context-Specific Description and Inference

Research Discussion. Lead Institution

Q&A / Discussion on the research

Announcement

Upcoming seminar in the series and other announcements


Event Recording