"Language Resources and Evaluation", special issue on wordnets and relations
Call for papers
The building of wordnets, a world-wide preoccupation now, comes with the inevitable fragmentation of effort, and with multiplicity of methods and underlying theories. It is not enough merely to translate WordNet (wordnet.princeton.edu/). Customization is required. Some teams have decided to steer altogether clear of this largest wordnet, so as better to reflect all the specificity of language structures and of culture. The Global Wordnet Grid (www.globalwordnet.org/gwa/gwa_grid.htm) is a still-rare attempt to bring some order into the fast-growing thicket of incompatible ideas.
Among the defining properties of a wordnet, perhaps the most central are the lexical-semantic relations which make up the net. Hypernymy and hyponymy are the backbone. Synonymy and antonymy are the cornerstone. In WordNet, there are four types each of meronymy and holonymy. Beyond that, WordNet has only a handful of relations.
The community has put a good deal of effort into adding instances to wordnet relations, notably to hypernymy and (to a lower degree) meronymy. Very seldom, however, do people question the relation list itself or evaluate relations on the basis of the data collected. Yet this is precisely what distinguishes wordnets, and -- more important -- languages which such wordnets are meant to model. There is an urgent need to assemble a body of research results, not absent thus far but not organized in a systematic manner either.
*
This is a call for papers to a special issue of the journal "Language Resources and Evaluation" , to appear early in 2012. It will group together work on wordnet and relations. The following topics will be of particular interest:
a) lexico-semantic relations in linguistics and in wordnets,
b) wordnet versus other types of thesauri, and relations therein,
c) the lexicographic theories and practices versus wordnet-creation practices,
d) mono-lingual and multi-lingual considerations in the creation of a wordnet,
e) the issues around translating a wordnet into another language,
f) comparing wordnets for one language and between languages from the standpoint of relation sets,
g) automatic extraction of lexical semantic relations and the role of large corpora in practical wordnet development,
h) evaluation of lexico-semantic relations in wordnets -- consistency, coverage, relevance for applications.
Papers supported by rich practical experience in large-scale wordnet development will be especially welcome: neat theories often fray at the adges when confronted with rich language data.
*
We invite original contributions, not published before and not under consideration for publication elsewhere. Each paper will be reviewed by two readers appointed by the journal's editors and two appointed by guest editors. After the initial review phase, the authors of accepted papers will have an opportunity to look at all sucessful submissions in order to extend their papers with elements of discussion and to cross-reference all contributions accordingly.
Important dates:
call for papers broadcast by October 31, 2010;
submissions due by May 31, 2011;
reviews to authors by September 30, 2011;
revisions of accepted papers due by November 30, 2011;
(in parallel) discussion between the authors of accepted papers till November 30, 2011;
final editorial decisions by December 31, 2011.
Reviewers for the special issue:
Irina Azarova, Saint-Petersburg University
Pushpak Bhattacharyya, Indian Institute of Technology
Francis Bond, Nanyang Technological University
Sonja Bosch, University of South Africa
Lars Borin, University of Gothenburg
Helge Dyvik, University of Bergen
Darja Fišer, University of Ljubljana
Chu-Ren Huang, Academia Sinica
Hitoshi Isahara, Toyohashi University of Technology
Kyoko Kanzaki, National Institute of Information and Communications Technology
Svetla Koeva, Bulgarian Academy of Sciences
Judit Kuti, Hungarian Academy of Sciences
Lothar Lemnitzer, Berlin-Brandenburg Academy of Sciences
Bernardo Magnini, FBK
Palmira Marrafa, University of Lisbon
Karel Pala, Masaryk University
Adam Pease, Articulate Software
Marko Tadić, Croatian Academy of Sciences and Arts
Dan Tufiş, Romanian Academy
Tamás Váradi, Hungarian Academy of Sciences
Piek Vossen, VU University Amsterdam
Guest editors:
Bolette Sandford Pedersen, University of Copenhagen
Maciej Piasecki, Wroclaw University of Technology
Stan Szpakowicz, University of Ottawa
Christiane Fellbaum, Princeton University