[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[eu_members at aclweb dot org] Colorado: NAACL HLT 2009 Workshop CFP: Active Learning for Natural Language Processing

              Call for Paper Submissions

                   NAACL HLT 2009 Workshop on
          Active Learning for Natural Language Processing

            June 5, 2009, Boulder, Colorado, USA

                 Submission Deadline: March 6, 2009

         Endorsed by the following ACL Special Interest Group:
    SIGANN, Special Interest Group for Annotation



Labeled data is a prerequisite for many popular algorithms in natural
language processing and machine learning.  While it is possible to obtain
large amounts of annotated data for well-studied languages in well-studied
domains and well-studied problems, labeled data are rarely available for
less common languages, domains, or problems.  Unfortunately, obtaining human
annotations for linguistic data is labor-intensive and typically the
costliest part of the acquisition of an annotated corpus.

It has been shown before that active learning can be employed to reduce
annotation costs but not at the expense of quality.  While diverse work over
the past decade has demonstrated the possible advantages of active learning
for corpus annotation and NLP applications, active learning is not widely
used in many ongoing data annotation tasks. Much of the machine learning
literature on the topic has focused on active learning for classification
problems with less attention devoted to the kinds of problems encountered in


We are interested in bringing together researchers in this area to explore
the challenges of active learning for NLP tasks.  General work on active
learning on NLP classification tasks, sequence labeling, parsing, semantics,
and other more complex tasks will be welcome in the workshop.  More specific
topics of interest include, but are not limited to:

  - theoretical analysis of active learning in the context of NLP
  - novel active learning approaches to estimate the training utility
    of individual selection units
  - cost-sensitive active learning approaches incorporating data
    acquisition costs
  - approaches to model or predict annotation costs as well as studies
    on factors that influence annotation time
  - criteria for stopping or monitoring progress of active learning
  - overfitting of data acquired with active learning: how much is the
    data biased towards the learning scheme involved in the selection
    and what are the limitations of re-use with other learning schemes
  - Human-Computer Interaction aspects of annotation including
    requirements, impact of interface design on annotation time, and
    methods to deal with reliability of annotators
  - approaches to multi-task active learning
  - approaches to deal with or reduce computational complexity of
    active learning approaches including parallelization, issues of
    pool- or batch-size, varying degrees of look-ahead, etc.
  - active learning and domain adaption
  - active learning compared to or combined with other semi-supervised
    or even unsupervised learning approaches
  - application of active learning in real annotation projects and
    experiences gained thereby


We invite submissions of two kinds:  1. original and unpublished work as
full papers, limited to 8 pages; 2. position papers or papers describing
ongoing work as short papers, limited to 4 pages.  Both kinds of papers will
appear in the proceedings and will be presented orally.  As reviewing will
be double-blind, author information should not be included in the papers and
self-reference should be avoided.

All submissions must be made in PDF format using the START paper submission
Submissions must follow the NAACL HLT 2009 formatting requirements:
Authors are strongly encouraged to use the LaTeX or Microsoft Word style
files available there.  Papers not conforming to these requirements are
subject to rejection without review.


March 6, 2009: Submission Deadline
March 30, 2009: Notification of acceptance
April 12, 2009: Camera-ready copies due
June 5, 2009: Workshop held in conjunctions with NAACL HLT


  - Eric Ringger, Brigham Young University, USA
  - Robbie Haertel, Brigham Young University, USA
  - Katrin Tomanek, University of Jena, Germany

Please address any queries regarding the workshop to:
              al dot nlp2009 at googlemail dot com


  - Shlomo Argamon (Illinois Institute of Technology, USA)
  - Jason Baldridge (University of Texas at Austin, USA)
  - Markus Becker (SPSS, UK)
  - Hal Daume (University of Utah, USA)
  - Robbie Haertel (Brigham Young University, USA)
  - Ben Hachey (University of Edinburgh, UK)
  - Udo Hahn (University of Jena, Germany)
  - Eric Horvitz (Microsoft Research, USA)
  - Rebecca Hwa (University of Pittsburgh, USA)
  - Ashish Kapoor (Microsoft Research, USA)
  - Mark Liberman (University of Pennsylvania/LDC, USA)
  - Ray Mooney (University of Texas at Austin, USA)
  - Miles Osborne (University of Edinburgh, UK)
  - Eric Ringger (Brigham Young University, USA)
  - Kevin Seppi (Brigham Young University, USA)
  - Burr Settles (University of Wisconsin, USA)
  - Katrin Tomanek (University of Jena, Germany)
  - Prem Melville (IBM T.J. Watson Research Center, USA)
  - Jingbo Zhu (Northeastern University, China)
  - Victor Sheng (University of Western Ontario, Canada)
  - Ken Church (Microsoft Research, USA)