[Corpora-List] JNLE Special Issue on Textual Entailment - Preliminary Announcement

[Dan Roth]

                 Journal of Natural Language Engineering
                   Special Issue on Textual Entailment

                Preliminary Announcement (CFP on April 2007)

The goal of identifying textual entailment - whether one piece of
text can be plausibly inferred from another - has emerged in recent
years as a generic core problem in Natural Language Understanding.
For instance, in order to answer the question "Who killed Kennedy?",
a QA system may need to recognize that "Oswald killed Kennedy" can
be inferred from "the assassination of Kennedy by Oswald".

Work in this area has been largely driven by the PASCAL Recognizing
Textual Entailment (RTE) challenges, a series of annual competitive
meetings (http://www.pascal-network.org/Challenges/RTE3). This work
exhibits strong ties to some earlier lines of research, particularly
automatic acquisition of paraphrases and lexical semantic
relationships and unsupervised inference in applications such as
question answering, information extraction and summarization. It
has also opened the way to newer lines of research on more involved
inference methods, on knowledge representations needed to support
this natural language understanding challenge and on the use of
learning methods in this context. RTE has fostered an active and
growing community of researchers focused on the problem of applied
entailment. The special issue of JNLE will provide an opportunity to
showcase some of the most important work in this emerging area.

The call for papers is anticipated in April 2007 with paper
submission deadline several months later. This schedule would allow
participants of the PASCAL RTE-3 Challenge to include their most recent
results, following the RTE-3 workshop. However, papers will be invited
on all aspects of textual entailment, aiming at a broader scope than
exhibited within the RTE challenges. Topics include, but are not
limited to:

* Representation levels, such as
    - Lexical, n-gram, and substring overlap
    - Linguistic annotations (POS tags, syntactic structure,
      semantic dependencies)
* Utilizing background knowledge, e.g. inference rules, paraphrase 
templates, lexical relations
* Knowledge acquisition methods
   - From corpora/Web, including construction of entailment/paraphrasing 
corpora
   - From semantic resources like FrameNet, PropBank, VerbNet, 
NOMLEX/NOMBANK
* Inference mechanisms, such as
    - Similarity/subsumption metrics
    - Tree-based distances and transformations
    - Machine learning
    - Logical inference using theorem provers
* The impact of entailment capabilities on applications
* Evaluation methods
* Data analysis

Guest Editors:
Ido Dagan          (Bar Ilan University, Israel)
Bill Dolan         (Microsoft Research, USA)
Bernardo Magnini   (ITC-irst, Italy)
Dan Roth           (UIUC, USA)