POLMETH Archives

Political Methodology Society

POLMETH@LISTSERV.WUSTL.EDU

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Katherine Keith <[log in to unmask]>
Reply To:
Political Methodology Society <[log in to unmask]>, Katherine Keith <[log in to unmask]>
Date:
Wed, 31 Jan 2024 11:10:51 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (158 lines)
Sixth Workshop on NLP and Computational Social Science (NLP+CSS) at NAACL
2024

Language is deeply intertwined with nearly all human social processes. We
do not expect teenagers to speak like senior citizens, and we recognize the
mutual dependency between language and the ways people interact with and
even conceptualize the world. Although this interdependence is at the core
of models in both natural language processing (NLP) and (computational)
social sciences (CSS), these two fields are continuing to come together,
with many opportunities for novel methods, research insights, and potential
applications. Humans with different social attributes and cultural
backgrounds (compared to bots and trolls) react to information spread
online differently, and express their reactions using a large variety of
language and content choices. Identifying and measuring bias based on
language use in different online communities is another emerging area of
research. Moreover, it has been shown that one can construct social
variables from language and estimate the relationship between these social
variables and measures in economics, politics, law, religion, anthropology
and other fields.

This workshop aims to (1) advance the joint computational analysis of
social sciences and language and (2) study how language can be used to
measure social variables and their impact across disciplines, both by
explicitly involving social scientists with NLP researchers, and other
partners from both industry and academia.

This sixth edition of the NLP+CSS workshop builds on five successful years
with hundreds of interdisciplinary submissions to make NLP techniques and
insights standard practice in CSS research. Our focus is on NLP for social
sciences: to continue the progress of CSS, and to integrate CSS with
current trends and techniques in NLP.

The workshop will have the following tentative format:

   -

   Invited speakers with a key emphasis on bringing in social scientists
   from outside NLP and industry participation,
   -

   Short talks for selected papers
   -

   A general poster session for all accepted papers


Submission Details

We invite research on any of the following general topics:

   -

   NLP models and data analytics that incorporate extra-linguistic social
   information
   -

   Development and/or application of NLP tools for computational social
   science problems
   -

   Methods or studies that test or revisit research from sociolinguistics
   -

   Approaches to identify bias based on language use in different
   communities
   -

   Insights into the importance of extra-linguistic attributes from NLP
   models across languages and cultures
   -

   Methods or applications that combine NLP with causal inference to better
   understand social-scientific processes
   -

   Use of large language models (LLMs) for social science measurement


Areas of interest include all levels of linguistic analysis and social
sciences, including (but not limited to): phonology, syntax, pragmatics,
stylistics, economics, psychology, sociology, sociolinguistics, political
science, geography, demography, survey methodology, and public health.

We especially invite graduate students from both disciplines (i.e. social
sciences and NLP) and connect them with experts in the respective other
field (e.g., an NLP student with an expert in social sciences or vice
versa). We would like to again provide mentorship for social science
students who could not otherwise attend a computer science conference.

Submission. We invite both long and short papers to be submitted through
Open Review:

https://openreview.net/group?id=aclweb.org/NAACL/2024/Workshop/NLP-CSS

Long papers should present new and substantial contributions related to the
workshop’s theme. Short papers may be a small and focused contribution or
describe a work in progress. While all submissions will be reviewed
equally, authors can choose a non-archival submission, since some social
sciences do not accept journal articles already published in archived
proceedings.

Papers will follow the <https://2022.emnlp.org/calls/style-and-formatting/>ACL
(ARR) style guidelines
<https://aclrollingreview.org/cfp#paper-submission-and-templates> for
length and formatting.

More Information

The NLP+CSS website contains more details on the call for papers,
submission instructions, and aims of the workshop:
https://sites.google.com/site/nlpandcss/ or follow us on Twitter/X at
@nlpandcss. <https://twitter.com/nlpandcss>

Contact the organizers by email via nlp-and-css [AT] googlegroups.com.

Important Dates

   -

   Workshop papers due: March 24th, 2024
   -

   Notification of acceptance: April 14th, 2024
   -

   Camera-ready papers due: April 24th 2024
   -

   Workshop dates (at NAACL in Mexico City): June 21 or 22, 2024

ORGANIZING COMMITTEE

   -

   Dallas Card (University of Michigan)
   -

   Anjalie Field (Johns Hopkins University)
   -

   Dirk Hovy (Bocconi University)
   -

   Katie Keith (Williams College)

**************************************************************
               Political Methodology E-Mail List
   Editors:  Yuan (Cecilia) Sui and Gechun Lin
                  <[log in to unmask]>
**************************************************************
     Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
                 your subscription settings visit:

  https://www.cambridge.org/core/membership/spm/mailing-list

**************************************************************

ATOM RSS1 RSS2