STEP2008 2nd CfP

delmont at unive.it delmont at unive.it
Sun Apr 27 12:16:16 UTC 2008


Apologies for Multiple Postings


  ----------------------------------------------------------------



                   2nd CALL FOR PAPERS: STEP 2008



                               P  2

                             E      0

                           T          0

                         S              8





              Symposium on Semantics in Text Processing



         http://project.cgm.unive.it/html/STEP2008/index.htm



                        September 22-24, 2008



                    Auditorium Santa Margherita

                          Venice (Italy)



         Endorsed by SIGSEM, the ACL special interest group

                   on computational semantics





 MOTIVATION

============

Thanks to both statistical approaches and finite state methods,

natural language processing (NLP), particularly in the area of robust,

open-domain text processing, has made considerable progress in the

last couple of decades. It is probably fair to say that NLP tools have

reached satisfactory performance at the level of syntactic processing,

be the output structures chunks, phrase structures, or dependency

graphs.  Therefore, the time seems ripe to extend the state-of-the-art

and consider deep semantic processing as a serious task in

wide-coverage NLP. This is a step that normally requires syntactic

parsing, as well as named entity recognition, anaphora resolution,

thematic role labelling and word sense disambiguation, as well as

other lower levels of processing for which reasonably good methods

have already been developed. Accurate automatic semantic

interpretation of text is expected to benefit newly emerging areas

targetting semantic and pragmatic issues, such as affectivity and

sentiment analysis of texts, textual entailment, and consistency

checking.





 WORKSHOP SCOPE

================

The goal of the STEP workshop is to provide a forum for anyone active

in semantic processing of text to discuss innovative technologies,

representation issues, inference techniques, prototype

implementations, and real applications. The preferred processing

targets are large quantities of texts -- either specialised domains,

or open domains such as newswire text, blogs, and wikipedia-like text.

Implemented rather than theoretical work is emphasised in STEP.

In particular, relevant topics are:



- wide-coverage semantic/logical analysis of text

- computation and use of discourse relations

- use of lexical-conceptual and semantically related resources

- thematic role labelling in semantic representations

- word sense disambiguation in semantic representations

- implementations of specific semantic phenomena

- anaphora or ellipsis resolution in semantic representations

- implementations of sentiment analysis

- automatic detection of subjective and non-literal language

- acquisition of lexical knowledge and paraphrase from raw corpora

- background knowledge acquisition, representation, and selection

- semantic lexicons and ontologies for text interpretation

- learning semantic representations from raw text

- automated reasoning in the service of semantic analysis of text

- creation of gold standard meaning representations

- evaluation of semantic representations

- textual entailment and consistency checking

- systems that extract, represent or manipulate text meaning

- applications of semantic analysis in text processing



Applications inlude, but are not limited to, machine translation, text

understanding, question answering, summarisation, information

extraction, and the semantic web.





 SHARED TASK: COMPARING SEMANTIC REPRESENTATIONS

=================================================

STEP 2008 will also feature a "shared task" to compare semantic

representations as output by state-of-the-art NLP systems.

Participating systems will be given a number of (small) texts, before

the workshop.  The output of these systems will be judged on a number

of aspects by a panel of experts in the field, during the workshop.

Aim of the shared task is to discuss the feasibility of a gold

standard for deep semantic representations. Aim of the panel is to

identify a set of problematic and relevant issues for semantic

evaluation.  The panel will reward the system with the most complete

and accurate semantic representation with a special prize. Preliminary

dates for the Step Shared Task are:



	Shared Task paper submission:     June  6, 2008

        Notification of acceptance:       June 23, 2008

        Release of test data              June 25, 2008

        System's results due              July  4, 2008

        Final version paper due:          July 25, 2008

        Workshop:                      Sept 22-24, 2008



To participate in the shared task, submit a paper containing (1) a

system description, (2) a description of the semantic formalism used

by the system, and (3) an authentic small text and the way it is

analysed by the system.  This text should not exceed five sentences

and 120 tokens. The test data for the shared task will be composed out

of all the texts submitted by the participants.



Shared task submissions should follow the workshop format for regular

papers and submission guidelines (see below), and will be published in

the STEP 2008 proceedings. Please mark shared task paper submissions

by specifying "shared task" as one of the keywords.  The final paper

must include a discussion of the system's performance on the shared

task data.  Please contact Johan Bos (bos at di.uniroma1.it) for further

questions on the shared task.





 SUBMISSIONS

=============

Authors are invited to submit original research papers. Papers should

indicate the state of completion of the reported results. Overlap with

previously published work should be clearly indicated.  Submissions

will be judged on correctness, novelty, technical strength, clarity of

presentation, significance, and relevance to the workshop.



Submissions should be in Abobe PDF format, not exceed eight A4-sized

pages, and be typeset in a 11 point font.  Detailed guidelines and a

latex stylefile package are available at the STEP 2008 web page. Paper

submission will be electronic using the EasyChair system.



Each submission will be reviewed by at least two members of the

programme committee. Accepted papers will be published in the workshop

proceedings. The publication of selected and revised papers is under

consideration for a special issue in a journal.





 INVITED SPEAKER

=================

        Harry Bunt (University of Tilburg)





 IMPORTANT DATES

=================

        Regular Paper submission deadline:  May 9, 2008

	Shared Task paper submission:      June 6, 2008

        Notification of acceptance:       June 23, 2008

        Camera-ready version due:         July 25, 2008

        Workshop:                      Sept 22-24, 2008





 ORGANISING COMMITTEE

======================

        Rodolfo Delmonte (Universita' Ca' Foscari, Venice)

        Johan Bos (Universita' La Sapienza, Rome)





 PROGRAMME COMMITTEE

=====================

        Roberto Basili (University Tor Vergata, Rome, Italy)

        Amedeo Cappelli (CELCT, Trento Italy)

        Ann Copestake (University of Cambridge, UK)

        Nicola Guarino (ISTC-CNR, Trento, Italy)

        Sanda Harabagiu (HLT, University of Texas, USA)

        Alexander Koller (University of Edinburgh, UK)

        Leonardo Lesmo (DI, University of Tourin, Italy)

        Katja Markert (University of Leeds, UK)

        Dan Moldovan (HLT, University of Texas, USA)

        Srini Narayanan (ICSI, Berkeley, USA)

        Sergei Nirenburg (University of Maryland, USA)

        Malvina Nissim (University of Bologna, Italy)

        Vincenzo Pallotta (Universitaet Freiburg, Schweiz)

        Emanuele Pianta (ITC, Trento, Italy)

        Massimo Poesio (University of Trento, Italy)

        Stephen Pulman (Oxford University, UK)

        Michael Schiehlen (IMS Stuttgart, Germany)

        Bonnie Webber (University of Edinburgh, UK)



More information about the LFG mailing list