Text as Data Association

In 2015, several members of the text as data community formed an association (the Text as Data Association) and a board to keep our annual “New Directions” conferences going.  The result has been a great series of meetings and a continually growing community.  At the end of 2018, our inaugural director, Arthur Spirling, passed the baton to me.  Many thanks to Arthur for getting the organization off to a great start.
The Text as Data association is concerned with the development and application of computational, quantitative methods for social scientific and humanistic inquiry into human behavior.  It is open to members from any field of inquiry, academic discipline, government or industry.  See https://textasdata.github.io for more about our goals and the list of the current board members.

A few updates:

  • Arthur and his team have updated the webpage:  https://textasdata.github.io
  • There will be another great conference this fall, at Stanford University, October 5-6; details will come soon
  • We will hold elections for some new board members in the next couple of months
  • We are now formalizing membership in the association, which is why I’m writing

To join the association for 2019, all you need to do is fill out this form:  https://forms.gle/mvreqkpPQJDWqa86A .  At this time, there is no cost for membership.
Please join so that our official database of members accurately reflects our community of living, breathing researchers!  (Also, so that you can nominate and vote for new board members.)  And please forward this message to anyone elsein the community who may not be reached via this mailing list.
Best regards,

Noah Smith

Director, Text as Data AssociationProfessor of Computer Science & Engineering, University of WashingtonSenior Research Manager, Allen Institute for Artificial Intelligence

Call for Papers!

Sixth Workshop on Computational Linguistics and Clinical Psychology: Reconciling Outcomes

NAACL 2019

June 6th

Minneapolis, MN

Web site: clpsych.org

A closer union between clinical psychology and natural language processing introduces a new frontier for mental health at a time when supporting more people —  and faster — is a financial and public health priority. There is a substantial and rising burden of mental health disorders and a significant shortage of resources available to prevent, diagnose, and treat them. For 5 years, CLPsych has brought together scientists and clinicians interested in improving the efficiency and reach of mental health treatment through scholarship and discussion focused on:

(a) objective, naturalistic behavioral data to improve upon the assessment of mental health needs provided by self-reported questionnaires — currently the most common method of assessment;

(b) providing language researchers a deeper understanding of the field of mental and neurological health and the needs of those currently responsible for diagnosis and treatment;

(c) increasing clinicians’ understanding of innovations in language technology and realistic expectations for implementation;

(d) the co-development of high-value NLP tools that can be deployed in the clinical community; and

(e) discussion of critical ethical questions concerning the opportunities and challenges created by human-technology interactions and analysis of the data generated.

CLPsych appeals to an interdisciplinary audience sharing findings, models, and methods to increase access to and scalability of mental health care.

This year’s workshop will focus on reconciling outcomes, a construct central to each of the main audiences participating in CLPsych; however, operationalized differently, with definitions ranging from treatment response to prediction accuracy. We encourage researchers to contribute papers that address different definitions of what constitutes an “outcome” in the intersection of language and clinical psychology, connected with focus areas (a-e) above. We are particularly interested in papers contributing to a discussion of how to align the outcomes that are most important to pursue as a community.

Paper submission instructions

A key goal of this workshop is to foster the conversation with clinicians, both at the workshop and when these papers are read in the future. We therefore include practicing clinicians and clinical researchers on our program committee, and the ability to communicate ideas, approaches, and results clearly to people who are not computational linguists will be as important as the quality of the work itself.  This year we will be taking two categories of submission:

Standard papers should follow the guidelines for Long Papers in the NAACL 2019 call for papers at https://naacl2019.org/calls/papers/.  These permit 8 pages of content plus references. In addition to papers describing algorithms, models, and experimentation, we are happy to receive carefully argued and supported position papers, insightful reviews or synthesis of relevant literature, or informative descriptions of real-world experiences deploying language technology (including prototypes) in relevant clinical settings.

Position papers can contain up to 4 pages of content, plus references of any length. These papers should clearly articulate a position, argument, or agenda that may be of value to the Workshop’s community. For example, authors with a clinical focus may have a position on ethical issues, promises and pitfalls of adopting technological approaches and digital mental health solutions, the connection between technology and interventions, or issues the CLPsych community should be considering with respect to specific conditions or populations. Position papers from either the technological or clinical communities are welcome and encouraged.

Submissions may be accepted for oral or poster presentation.  All papers must conform to the Author Guidelines at https://naacl2019.org/calls/papers/.  Please make sure to review NAACL’s recently updated policies regarding anonymity and double-blind review.

Important Dates

6 March 2019: Papers due (https://www.softconf.com/naacl2019/clpsych/)

27 March 2019: Notification of acceptance

5 April 2019:  Camera-ready papers due

6 June 2019:  Workshop!

Shared task. The workshop will include a shared task which will be announced separately in February 2019 with its own important dates.

Organizers (clpsych-organizers@googlegroups.com)

Kristy Hollingshead, IHMC

Kate Loveys, University of Auckland School of Medicine

Kate Niederhoffer, 7 Cups

Philip Resnik, University of Maryland

Rebecca Resnik, Rebecca Resnik and Associates, LLC

Workshop site: http://clpsych.org/NAACL conference site: http://naacl2019.org/

Save the date: PaCSS 2019

I’m pleased to announce that the 2nd Annual Politics and Computational Social Science (PaCSS) conference will be held at Georgetown University on Wednesday August 28, 2019. 

Initial information is available at https://mccourt.georgetown.edu/PaCSS.

We will provide updates on application process later at the website and via email.  Please feel free to email me atMichael.Bailey@georgetown.edu if you have questions.

If you’re not familiar with the conference, here is a summary: The data and methodologies available to social scientists have exploded with the emergence of vast archives of passive data collection, large scale online experimentation, and innovative uses of simulation. These data are of a larger magnitude and methods are of a greater computational complexity than approaches that have dominated political science for the last 50 years. This offers the potential for rich insights into society at scale, while simultaneously introducing new ethical and infrastructural challenges. In parallel, the information and communication technologies that have driven these changes are also driving changes in politics, around the world, that require study. In order to understand the political world, it is increasingly important to gain access to the political communication and behavior occurring online. This conference gathering will offer a forum for political science research in this emerging space. Example relevant topics/approaches include:  analysis of social media; text analysis; use of finely granular geographic data; large scale online experimentation. Scholars from all subfields of political science are welcome, as are researchers in other relevant disciplines, such as computer science, sociology, and physics.

PolText Conference

CALL FOR PAPERS

POLTEXT 2019: The 3rd International Interdisciplinary Symposium on the Quantitative Analysis of Textual Data

14-15 September 2019, Waseda University, Tokyo, Japan

The next POLTEXT symposium (https://www.poltextconference.org) will take place in Tokyo, Japan in September 2019. The conference aims to facilitate the knowledge exchange between social scientists and computer scientists who address social, political or economic issues through the analysis of textual data by employing quantitative methods. It also features multi-lingual analysis of textual data to promote the methodology beyond the English language. In this, the symposium will be the first of its kind in Japan. Jonathan Slapin (University of Essex) and Amy Catalinac (New York University) have been confirmed as keynote speakers related to these two themes, respectively.

We strongly encourage scholars who employ quantitative text analysis methods to submit abstracts for presentation at the conference. Both substantive and methodological papers will be accepted for presentation. Substantive papers may be related to any social science or humanities discipline as long as the project is based on a quantitative analysis of textual data. Possible subjects for methodological papers may include text classification, topic modeling, information extraction, sentiment analysis, ideal point estimation, argumentation mining, political reputation analysis, and multilingual comparison.

Following the precedence of the previous POLTEXT Incubator Workshop in 2018 (https://poltext.tk.mta.hu/en), we are planning to offer free tutorial sessions (e.g. Quantitative Text Analysis in R, Quantitative Analysis of Textual Data using Machine Learning) to participants on the day before the main event (13 September). If you are interested in offering a tutorial on methods or text mining tools, please contact the organizers.

The symposium is organized by Kohei Watanabe (Institute for Advanced Study, Waseda University), Miklós Sebők (Centre for Social Sciences, Hungarian Academy of Sciences) and Lisa Lechner (University of Innsbruck).

SUBMISSION OF PROPOSALS AND PAPERS

Proposals, along with an abstract of 250 words and six keywords, should be submitted via the submission form (https://www.poltextconference.org/submit-an-abstract) by 31 March 2019. Notifications of acceptance are due by 30 April. The registration deadline is 30 August. Full-length papers must be uploaded by 31 August. Selected papers will be published in the symposium proceedings.

TRAVEL GRANTS FOR EARLY CAREER PRESENTERS

We understand that traveling to Japan is a costly affair. Therefore our aim is to offer financial support to young scholars traveling from overseas with the help of charitable foundations and sponsorship by Japanese scholars. If you need financial support, please indicate in the abstract submission form (https://www.poltextconference.org/submit-an-abstract). Full or partial travel funds will be available for graduate students and early career researchers (up to four years since doctorate has been granted) who are willing to take part in bi-lateral co-operations with Japanese scholars and host institutions. We will offer more detail on these travel options as soon as funds become available.

Conference on Language Endangerment and Political Instability

Political instability and violence create the environment for language loss and resulting social instability. Establishing the intersecting causes and consequences of political and social instability as tied to language is of interest to national security. The purpose of the Political Instability and Language Endangerment Conference is to develop a thorough understanding of how political instability and violence break apart communities and erode their ability to pass down their written and spoken word. This interdisciplinary conference will bring together experts from political science, linguistics and other fields to develop a global and comprehensive account of the intersection of politics and language endangerment. Broader impacts include the engagement of undergraduate and graduate students, many of whom are first generation and themselves from underrepresented groups, by providing networking and co-authoring opportunities to mentor students and to provide training that will be invaluable in both private and public sector work.

Led by an interdisciplinary team of linguists and political scientists from the University of North Texas, the Political Instability and Language Endangerment Conference will integrate knowledge from both disciplines and will produce new strategies for interdisciplinary research. The dominant paradigm for understanding the causes of language endangerment takes into consideration local social and economic factors. The potential for a more comprehensive understanding is possible through the perspective of political science which takes into account the role of violence, power and resources in societies and across institutions of governance. In order to stimulate research utilizing this novel interdisciplinary perspective, conference participants will explore (1) how government policy, instability due to war and other forms of conflict lead to language endangerment; (2) how instability due to economically forced immigration contributes to language endangerment; and (3) how political conflict due to language endangerment and national language policies can lead to political protest, unrest and violence. The resulting white paper and edited volume will be invaluable resources for practitioners, academics, and most importantly, the affected communities to develop strategies for language preservation. A follow-up conference envisioned for the Smithsonian Institution will help raise greater public awareness of this phenomenon.

NYU Text-as-Data Fall Speakers Series

The NYU NLP and Text-as-Data Speaker Series takes place on Thursdays from 4 – 5:30 pm at the Center for Data Science, 60 Fifth Avenue (7th floor common area).  Expanding from its original emphasis on text data applied to social science applications, the Series incorporates the growing interest in Natural Language Processing from a variety of disciplines, especially Computer Science and Linguistics. The series provides an opportunity for attendees to see cutting-edge NLP and other text-as-data work from the fields of social science, computer science and other related disciplines.  The seminar is organized by Professors Arthur Spirling and Sam Bowman, and is open to anyone who is affiliated with NYU and wishes to attend.

PaCSS keynote speech (Gary King)

The keynote speech for the Preconference on Politics and Computational Social Science is available here. A roundup of the best PaCSS tweets here.

Ahead of this year’s APSA general meeting, we attended the Politics and Computational Social Science (PaCSS) pre-conference, hosted at Northeastern University. The event brought together  political scientists working with large-scale data sets and emerging computational methods.