LACSS 2014 - Workshop on Language Technologies and Computational Social Science

A major growth area in applied natural language processing has been the field of computational social science, in which automated techniques are applied to massive datasets to answer scientific questions about society. Although much work in computational social science focuses on structured data or network data, linguistic data is also central. While some existing natural language processing techniques have found use in this growing community, new techniques for discovering and analyzing social meanings and structures in text are in high demand.
Tackling these challenges should be an interdisciplinary pursuit, building on expertise not just in language technologies but also in substantive social science fields (e.g., political science, economics, sociology, etc.). In particular, engagement between NLP researchers and social scientists will introduce new problem formulations and new theoretical frameworks that will broaden and deepen applications of language technology to social science.
The goal of this workshop is to increase the visibility of this application area for ACL researchers and to help build connections between language technologists and social scientists. The workshop is organized around invited talks from researchers who have successfully brought language technologies to computational social science research questions. Following each invited talk session, there will be an open discussion period.
We also invite abstracts of research in progress relevant to the theme of language technologies for computational social science. Work in progress is encouraged, and all presentations will be presented as posters. This format aims at fostering interactions among participants and invited speakers, contributing towards building a community interested in language technologies and computational social science.
Unshared task:
In coordination with the NSF-funded PoliInformatics research coordination network, the workshop will host a session at which the results of the PoliInformatics NLP “unshared task” will be announced. The unshared task is being independently organized by Noah Smith, Claire Cardie, and the PoliInformatics principal investigators (John Wilkerson and Anne Washington). It is an open challenge for NLP researchers to help make sense of U.S. legislative data relating to the financial crisis; entries in the task will be evaluated by a panel of social scientists, with the hope of matching techniques with specific data analysis problems. The session at this workshop is a culminating event for the challenge that seeks to bring together participants for discussion and increase exposure of the work.