MWE Workshop on Lexicography and NLP Perspectives
Explore the intersection of lexicography and Natural Language Processing (NLP) at the MWE workshop in Skopje, FYROM. Discover how lexicons can enhance NLP tools, and draft specifications for lexicographic software and MWE lexicons usable in NLP applications. Engage with experts in the field and contribute to the joint publication based on the workshop results.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
PARSEME/ENeL workshop on MWE e-lexicons Skopje, FYROM
Invitation (10 Dec 2015) ENeL has been collaborating closely with the Parseme COST action, PARSing and Multi-word Expressions The objective is to look for synergies between our two communities, and in particular to: understand how lexicographers organize MWEs in dictionaries and lexicons and what their software needs are understand how NLP tools and applications can benefit from MWE lexicons, draft specifications of an ideal lexicographic software, draft specifications of an ideal MWE lexicon usable in NLP applications. The drafted specifications will be made publicly available. We are also planning a joint publication based on the results of the workshop.
Practical information Dates: 5-6 April 2016 (co-located with PARSEME's 6th general meeting on 7-8 April in Struga) Location & Hosting Institution: University of Skopje, Faculty of Computer Science and Engineering (FCSE), Skopje, FYR of Macedonia (Katarina Zdravkova) Organizers: Simon Krek, Carole Tiberius, Carla Parra Escart n, Agata Savary Web site: http://typo.uni-konstanz.de/parseme/index.php/2- general/135-enel-parseme-workshop-on-mwe-lexicons
Tuesday morning 9:00-9:30 Presentation Welcome by workshop leaders and presentation round (max. 1 min. per participant) 9:30-10:30 Session 1: MWEs from a lexicographical perspective - a global view Construction of MWE lexicographical resources. Presentation of frameworks and standards used in Lexicography. Some demos of how lexicographers work Invited speakers: Lut Colman, Institute for Dutch Lexicology, the Netherlands Polona Gantar, University of Ljubljana, Slovenia 11:00- 12:00 Session 2: MWEs from a Natural Language Processing perspective - a global view Brief introduction to some NLP tools and applications. How can MWE lexicons improve them? How can NLP tools help in MWE lexicon construction? Invited speaker: H ctor Mart nez Alonso, University of Paris 7 - INRIA (France)
MWEs from a lexicographical MWEs from a lexicographical perspective perspective - - a global view a global view Lut Colman, Institute for Dutch Lexicology, the Netherlands Polona Gantar, University of Ljubljana, Slovenia
Introduction MWE typology from the dictionary user, lexicographer and NLP researcher perspective MWEs as parts of dictionaries and the mental lexicon Overview of different types of MWEs in dictionaries Properties of MWEs crucial for identifying different MWE types
Outline Lexicographic definition of MWE Concepts collocability, contiguity, idiomaticity, compositionality, figuration, fixedness Theoretical frames Treatment of MWEs in dictionaries macrostructure & microstructure description Semantics: defined or not, additional info (examples, labels, etc) Internal structure
Tuesday afternoon 13:00- 15:00 Session 3: PARSEME meets ENeL Plenary interactive demos of MWE resources and tools of the ENeL community Sketch engine, presented by Milo Jakubi ek MWE dictionaries and more: practical session by Lut Colman and Polona Gantar 15:30- 18:00 Session 4: ENeL meets PARSEME Plenary interactive demos of MWE-aware NLP tools of the PARSEME community mwetoolkit, presented by Carlos Ramisch LeXimir, presented by Cvetana Krstev Unitex, presented by Matthieu Constant Demo session of MWE-aware NLP tools
Data sets from participants Idion (Modern Greek), Stella Markantonatou (ILSP, Athens) - http://www.idion.ilsp.gr Polytropon (Modern Greek), Voula Giouli (Institute for Language and Speech Processing) - http://goo.gl/SgQlxS at http://athena.clarin.gr/ Kollokationenw rterbuch (Swiss Geman), Tobias Roth (Schweizerisches Idiotikon , Z rich) - http://www.kollokationenwoerterbuch.ch/web/ ANW (Dutch), Lut Colman (INL, Leiden) - http://anw.inl.nl Kamusi (different languages), Martin Benjamin (Kamusi.org) - https://kamusi.org/ Elhuyar MWE resources (Basque), Antton Gurrutxaga (Elhuyar Foundation, Usurbil) https://www.elhuyar.eus/en/site/community/nor-gara-en/fundazioa-en Valency Database of Croatian, Matea Birti (Institute for Croatian Language and Linguistics, Zagreb) - http://ihjj.hr/projekt/baza-hrvatskih-glagolskih-valencija/27/ Slovene Lexical Database, Polona Gantar (University of Ljubljana) http://eng.slovenscina.eu/spletni-slovar/leksikalna-baza
Wednesday 9:00-10:30 Session 5: Towards ENeL/PARSEME synergies - hands-on work I Working in groups with people from both communities Drafting specifications of an ideal NLP-supported lexicographic framework Designing a prototype interface with requested functionalities 15:30- 18:00 Session 6: Towards PARSEME/ENeL synergies - hands-on work II Working in groups with people from both communities Drafting specifications of an ideal MWE resource that would be useful in different NLP applications Sample data sets with the desired properties 13:30- 15:00 Final discussion and conclusions Comparing specifications from sessions I and II and discussing the reasons of divergences Planning a joint publication
Wish list One of the goals of the workshop is to draft specifications of an ideal NLP- supported lexicographic framework including requested functionalities dedicated to MWEs. In order to have a starting point for the work and discussions to be held during the workshop, we ask you to submit a wish list for an ideal NLP- supported lexicographic framework with requested functionalities. This wish list may include: ideal functionalities of MWE tools used for lexicographic purposes prototype interface for MWE tools used for lexicographic purposes description how to include MWE tools in lexicographic workflow other ideas how to use MWE tools for lexicography
ENeL Participants Simon Krek, Slovenia Carole Tiberius, Netherlands Lut Colman, Netherlands Polona Gantar, Slovenija Milo Jakubi ek, Czech Republic Tobias Roth, Switzerland Martin Benjamin, Switzerland Matea Birti , Croatia Thierry Declerck, Germany Antton Gurrutxaga, Spain Stella Markantonatou, Greece Voula Giouli, Greece Lidija Tanturovska, Macedonia Biljana Nikovska, Macedonia
Results Reading lists (ENeL Parseme) Data sets from participants (ENeL) Tools from participants (Parseme) Specifications of an ideal NLP-supported lexicographic framework (Prototype interface with requested functionalities) Specifications of an ideal MWE resource that would be useful in different NLP applications Sample data sets with the desired properties Joint publication
