Apologies for cross-postings
----------------------------------------
*** EXTENDED DEADLINE: Call for Journal Articles ***
Special Issue: LDV Forum 2008, Vol. 23(2)
Title: "Automatic Genre Identification: State of the Art, Issues, and Prospects"
*********************************************************
EXTENDED DEADLINE: 14 APRIL 2008
*********************************************************
RATIONALE
The popularity of the World Wide Web has given rise to distinct forms
of information-seeking behaviour and new interaction styles, but we
have not fully understood what these forms are nor their implication
for the developmen of new applications to assist users in the tasks
they perform online on a day-to-day basis. We believe that the concept
of genre have a significant impact on the future of web information
seeking and interaction, and our common goal is to explore the
possibility of meeting users' needs by filtering or organizing
information by genre.
There are several reasons why Automatic Genre Identification is still
only a niche research area. Among others, it seems to be very
difficult to agree upon a shared definition of genre, and a shared set
of genre categories. Another major problem is the lack of a genre
benchmark to perform objective evaluations of categorisation
approaches.
From the point of view of genre theory, the study on genres on the web
probes into the grey area at the crossing between semantics,
pragmatics, and information studies. The way in which people search
for genres on the web is very little explored, and only preliminary
investigations are available so far.
Regardless all these difficulties, 2007 has been an important year for
research on genres of digital documents. In addition to the genre
minitrack at the annual Hawaii' International Conference on System
Sciences (HICSS, usually held in January), three other events have
added findings to genre research, namely the Colloquium "Towards a
Reference Corpus of Web Genres", held in conjunction with Corpus
Linguistics 2007 in Birmingham (UK), the Workshop "Towards
Genre-Enabled Search Engines: The Impact of NLP" held in conjunction
with RANLP 2007, at Bovorets (Bulgaria), and finally the Panel
"Towards the Use of Genre to Improve Search in Digital Libraries:
Where Do We Go from Here?" sponsored by SIG-CR and SIG-HCI and held in
conjunction with ASIST 2007 Milwaukee, Wisconsin (USA).
This LDV special issue on genre originates from the organisers of the
colloquium and the workshop mentioned in the previous paragraph.
Contributors to these events are encouraged to submit fully-revised
and extended articles based on their workshop papers or colloquium
abstracts, enhanced with additional findings and more in-depth
analyses.
***We also encourage other researchers to submit articles to the
special issue.***
The main themes that we would like to focus on this issue are:
* the construction of reference corpora of web genres
* the inclusion of genre classes in IR systems and automatic document
categorisation
* investigations on users' behavior with regards to searching by genre
CONTACT INFORMATION
For questions or comments, please contact the editors at
webgenres([at])gmail([dot])com.
GUEST EDITORS: M. Santini, G. Rehm, S. Sharoff, A. Mehler
SCHEDULE
Extended Deadline: 14 April 2008
Notification to authors: 1 June 2008
Final versions due: 30 June 2008
Publication: Second half of 2008
SUBMISSION GUIDELINES
All submissions will be peer-reviewed following the double-blind
review process of LDV. Please, use the style sheet available from
http://www.ldv-forum.org/index.php?modus=style_sheets&language=en.
Send your submission to webgenres([at])gmail([dot])com and specify
"Submission to LDV Forum 2008, Vol. 23(2)" in the Subject line.
ABOUT LDV
LDV is short for Linguistische Datenverarbeitung (i.e. linguistic data
processing).
The LDV-Forum offers a publication platform for all researchers
working in the field of computational linguistics and language
technology. LDV is an open access journal and can accessed at
http://www.ldv-forum.org/index.php?modus=home&language=en.
|
