September 19, 2017: Domain Adaptation Techniques Workshop

September 19, 2017: Domain Adaptation Techniques Workshop

With the objective of better understanding recent advances and proposals in the field of adaptable NLP tools, TUNER organizes a workshop on Domain adaptation techniques to be held the 19th of September 2017, collocated with the XXXIII edition of the annual International Conference of the Spanish Society for Natural Language Processing (SEPLN) which will take place at the University of Murcia (Spain) on 20-22 September 2017 (http://sepln2017.um.es).

The workshop aims to bring together researchers and developers interested in Domain Adaptation Techniques in all branches of Natural Language Processing and compile information about current lines of research. We welcome papers describing current research and envisaged solutions, challenges, open problems and actual needs of industrial systems.

Accepted submissions will be presented in a central poster session followed by a panel discussion session where representatives of industry, public administration and other potential users of NLP-based applications will present their desiderata.

Workshop TUNER on Domain Adaptation Techniques for NLP tools

The success of the industrialization of text analytics and related big data applications depends on the quality of the results of text-processing tools. However, while industry and public administration need to process text of different genres and domains, most of the best NLP tools have been trained on a limited amount of texts, typically news. When these tools are used in other domains, a dramatic decrease in performance affects the credibility of the capabilities of these technologies.

Usually, domain data, annotated for different tasks, is required for every new application domain. For the text analytics providers in general, and SMEs in particular, this means that a significant amount of resources need to be devoted to manually annotating training data whenever new text genres or application domains are to be targeted. This challenge limits the expansion of companies to emerging business opportunities. Furthermore, it impedes the adoption of NLP technology by the public administration for important applications. Examples of these are clinical decision support or the automatic monitoring of tourist's opinion.

The TUNER Project (funded by the Spanish Ministerio de Economía, Comercio y Competitividad) aims to address the research and development of domain adaptation techniques that help the industrialization process of NLP-based applications.

With the objective of better understanding recent advances and proposals in the field of adaptable NLP tools, TUNER organizes a workshop on "Domain adaptation techniques" to be held the 19th of September 2017, collocated with the XXXIII edition of the annual International Conference of the Spanish Society for Natural Language Processing (SEPLN) which will take place at the University of Murcia (Spain) on 20-22 September 2017 (http://sepln2017.um.es).

The workshop aims to bring together researchers and developers interested in Domain Adaptation Techniques in all branches of Natural Language Processing and compile information about current lines of research. We welcome papers describing current research and envisaged solutions, challenges, open problems and actual needs of industrial systems.

Accepted submissions will be presented in a central poster session followed by a panel discussion session where representatives of industry, public administration and other potential users of NLP-based applications will present their desiderata.

Poster Abstract Submission

Abstracts for posters should be up to 2 pages long plus references. Submissions should adhere to the SEPLN 2017 format (style files available http://www.sepln.org/home-3/journal/author-guidelines/?lang=en&lang=en), and should be in PDF format.

Abstract submissions should be send to: nuria.bel@upf.edu

Important dates

14-05-2017: Workshop Poster Abstracts Due Date

14-06-2017: Notification of Acceptance

14-07-2017: Camera-ready Abstracts Due

September 19, 2017: Domain Adaptation Techniques Workshop

Program Committee

  • Rodrigo Agerri (Universidad del País Vasco)
  • Horacio Saggion (Universidad Pompeu Fabra)
  • Eneko Agirre (Universidad del País Vasco)
  • Xavier Gómez Guinovart (Universidad de Vigo)
  • Antoni Oliver (Universitat Oberta de Catalunya)
  • Eli Comelles (Universitat de Barcelona)
  • Montse Cuadros (Vicomtech)
  • Anselmo Peñas (Universidad Nacional de Educación a Distancia)
  • Iñaki San Vicente (Elhuyar)

Workshop organizers

German Rigau (Universidad del País Vasco)

Núria Bel (Universidad Pompeu Fabra)