This workshop focuses on the implementation of software and algorithms used in Natural Language and Speech Processing. Authors are invited to provide detailed descriptions, comparisons and/or evaluation of software algorithm implementations in any area of computational linguistics. The goal is to bring together researchers, software developers, teachers, and students with a common interest in the implementation of NLP applications, and to allow useful implementation techniques and "tricks of the trade" to be discussed in detail and disseminated widely.
Descriptions of user interfaces or end-to-end systems are not appropriate for this workshop and should be submitted to the Demo Session instead. Papers about novel algorithms are welcome, but must also describe an implementation. Descriptions of toolkits or libraries are discouraged, unless their main focus is on algorithmic and implementational aspects. Authors of accepted workshop papers are encouraged to include practical demonstrations in their presentations; however, the main aim of such presentations must be to demonstrate the internal workings of a particular software artefact to other researchers.
Relevant areas of interest include (without limitation): pragmatics, discourse, semantics, syntax, grammars and the lexicon; phonetics, phonology and morphology; lexical semantics and ontologies; word segmentation, morphological analysis, tagging and chunking; parsing, generation and summarization; language modeling, spoken language recognition and understanding; linguistic, psychological and mathematical models of language; language-oriented information retrieval, question answering, and information extraction; machine learning for natural language; corpus-based modeling of language, discourse and dialogue; multi-lingual processing, machine translation and translation aids; multi-modal and natural language interfaces and dialogue systems.
Possible topics for workshop papers include:
Papers should describe original, unpublished research. Reports on work in progress that meet all other submission criteria are welcome. Authors of accepted papers discussing software artefacts are required to make their implementations available to the research community under reasonable conditions by the time of the workshop. There will be an opportunity to distribute software on the Proceedings CD; authors of accepted papers are strongly encouraged to make use of this.
The proceedings will appear in electronic form only (CDs for registered workshop participants, ACL Anthology after the workshop). Authors of accepted papers are required to register for the workshop and will give 30 minute presentations, which can include practical demonstrations.
Initial submissions should consist of no more than 8 pages of text, plus an unlimited number of pages for references, large figures, code listings, long tables, and other appendices. Final papers may be longer, subject to file size limitations (details to be determined). Submissions should follow standard ACL procedures in all other respects: in particular, they must be anonymous and formatted in accordance with the official ACL style sheets.
This workshop accepts only electronic submissions in PDF format. Submissions should be sent by email as follows:
To: Martin Jansche <jansche {at} cs.columbia.edu>
Subject: ACL 2005 Workshop on Software submission
Date: Sun, 24 Apr 2005 23:59:59 -0500
In the body of the email, please indicate the names and affiliations of all authors and the email address of a contact person (if different from the submitter). Also indicate in the body of the email if the paper has been or will be submitted to any other conferences. Attach a PDF file of your submission to the message. No identifying information may be present in the PDF file.
Reviewing will be blind. Papers will be judged on technical merit, originality, significance, and relevance to the workshop.
See here.
Please direct any questions about this workshop to Martin Jansche <jansche {at} cs.columbia.edu>.