The Center for Social Science Research (CSSR) invites faculty, research staff, and students – especially those within the social sciences, including the College of Arts and Letters, the Law School, the Mendoza College of Business and the Keough School of Global Affairs – to attend a research talk by Professor Arthur Spirling (New York University) on Thursday, February 20, 2020 from 12:00 to 1:30 p.m. in 1030 Jenkins Nanovic Halls. (Flyer)
Professor Spirling’s talk is titled “Word Embeddings: What Works, What Doesn’t, and How to Tell the Difference for Applied Research”
ABSTRACT:
Word embeddings are becoming popular for political science research, yet we know little about their properties and performance. To help scholars seeking to use these techniques, we explore the effects of key parameter choices—-including context window length, embedding vector dimensions and pre-trained vs locally fit variants—-on the efficiency and quality of inferences possible with these models. Reassuringly we show that results are generally robust to such choices for political corpora of various sizes and in various languages. Beyond reporting extensive technical findings, we provide a novel crowdsourced “Turing test”-style method for examining the relative performance of any two models that produce substantive, text-based outputs. Our results are encouraging: popular, easily available pre-trained embeddings perform at a level close to—-or surpassing—-both human coders and more complicated locally-fit models. For completeness, we provide best practice advice for cases where local fitting is required.
BIO:
Arthur Spirling is Professor of Politics and Data Science at New York University. He received a bachelor’s and master’s degree from the London School of Economics, and a master’s degree and PhD from the University of Rochester. Spirling’s research centers on quantitative methods for social science, especially those that use text as data and more recently, deep learning and embedding representations. His work on these subjects has appeared in outlets such as the American Political Science Review, the American Journal of Political Science, the Journal of the American Statistical Association and conference proceedings in computer science. Substantively, he is interested in the political development of institutions, especially for the United Kingdom. His work has been supported by grants from the National Science Foundation for data infrastructure and training.
Food and refreshments will be available. To learn about this event and the CSSR, please visit cssr.nd.edu.
The Center for Social Science Research, which is a part of the Center for Research Computing, is dedicated to improving the quality and efficiency of social science research at the University of Notre Dame. The center offers computational services to faculty, students, and staff and offers expertise in statistical analysis, survey design and implementation, data acquisition and management, geographic information system (GIS) mapping, and more. To learn more about the center, please visit cssr.nd.edu.
Contact:
Brian Fogarty / Director
Notre Dame Research / University of Notre Dame
bfogart2@nd.edu / 574.631.6166
About Notre Dame Research:
The University of Notre Dame is a private research and teaching university inspired by its Catholic mission. Located in South Bend, Indiana, its researchers are advancing human understanding through research, scholarship, education, and creative endeavor in order to be a repository for knowledge and a powerful means for doing good in the world. For more information, please see research.nd.edu or @UNDResearch.
Originally published by at cssr.nd.edu on January 27, 2020.