Wordnets are increasingly widely used to model word meaning in natural language processing tasks. However, there are still many challenges in accurately describing word meanings and making these descriptions useful for both human and machine users. This workshop aims to identify, discuss and start to solve existing challenges.
The program for this workshop will be a little unusual. First, we will ask participants to read the papers ahead of the workshop. Then the workshop itself will be split into two sessions.
In the first session, each participant will have 15 minutes to highlight one or two of the challenges they consider most important, possibly with an outline of a solution. This will be followed by the keynote speaker, Ed Hovy, presenting some challenges he identified within his work.
In the second session, we will have a long moderated discussion with all participating. Here we try to come up with a deeper understanding of the problems and sketch some possible solutions.
|13:15–13:30||How Stable are WordNet Synsets?||Eric Kafe|
|13:30–13:45||Inside Baseball: Coverage, quality, and culture in the Global WordNet||Martin Benjamin|
|13:45–14:00||Two corpus based experiments with the Portuguese and English Wordnets||Fabricio Chalub, Alexandre Rademaker and Cláudia Freitas|
|14:00–14:15||Overview and Future of Czech Wordnet||Adam Rambousek, Karel Pala and Sandra Tukacova|
|14:15–14:30||Challenges behind the data-driven Bulgarian WordNet (BulTreeBank Bulgarian Wordnet)||Petya Osenova and Kiril Simov|
|14:30–14:45||The Revision History of Estonian Wordnet||Neeme Kahusk and Kadri Vider|
|14:45–15:00||Towards Revised System of Verb Wordnet Relations for Polish||Agnieszka Dziob, Maciej Piasecki, Marek Maziarz, Justyna Wieczorek and Marta Dobrowolska-Pigoń|
|15:00–15:15||Classification of Adjectives in BulNet: Notes on an Effort||Tsvetana Dimitrova and Valentina Stefanova|
|15:15–15:30||The Concept of Lexical Platform||Maciej Piasecki, Tomasz Walkowiak, Ewa Rudnicka and Tomasz Naskręt|
|16:00–16:30||Challenges for Wordnets (Keynote)||Eduard Hovy|
|16:30–16:40||Agenda Setting||All, Francis Bond (Moderator)|
|18:40–19:00||Conclusions, thank yous and farewells||Francis Bond and Maciej Piasecki|
The workshop is supported by the CLARIN-PL research infrastructure.
LDK 2017 conference received support from the Global WordNet Association Board.
Much research on wordnets focuses either on their construction or their use in some application. Few papers bridge the gap by discussing how different wordnet models and construction methods affect their effectiveness in use, or how different applications require different parts of language to be modeled.
For this reason, we are experimenting with a new workshop series challenges for wordnet, to give wordnet users and developers a chance to share experiences both good and bad. It will be co-located with the First Conference on Language, Data and Knowledge (LDK 2017) in Galway, Ireland. The workshop will start with some short presentations and then finish with an extended discussion based on the challenges presented. We welcome position statements on the following (or related) topics:
Please submit papers of between 6-10 pages, excluding references, formatted using the Springer Lecture Notes in Artificial Intelligence formatting guidelines. Submissions should be anonymous. Submissions will be reviewed by at least 3 reviewers and will be made available on online prior to the workshop.
Authors of good submisions will be invited to submit extended versions for a special issue of the Cognitive Studies | Études cognitives journal (https://ispan.waw.pl/journals/index.php/cs-ec/index). The extended versions will be carefully peer reviewed, but the scope of this special issue will be set in advance.
Papers should be submitted via EasyChair
Address any questions to Maciej Piasecki and Francis Bond at <firstname.lastname@example.org>.