[ Skip to the content ]

Institute of Formal and Applied Linguistics

at Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic


[ Back to the navigation ]

Publication


Year 2012
Type in proceedings
Status published
Language English
Author(s) Cinková, Silvie Holub, Martin Kríž, Vincent
Title Optimizing semantic granularity for NLP - report on a lexicographic experiment
Czech title Optimalizace sémantické granularity pro NLP - zpráva o jednom lexikografickém experimentu
Proceedings 2012: Oslo, Norway: EURALEX 2012: Proceedings of the 15th EURALEX International Congress
Pages range 523-531
How published data storage
Supported by 2010-2015 LM2010013 (LINDAT-CLARIN: Institut pro analýzu, zpracování a distribuci lingvistických dat) 2010-2013 GAP406/10/0875 (Komputační lingvistika: Explicitní popis jazyka a anotovaná data se zřetelem na češtinu) 2012-2018 GBP103/12/G084 (Centrum pro multi-modální interpretaci dat velkého rozsahu) 2012-2016 PRVOUK P46 (Informatika)
Czech abstract Experimenty se sémantickou anotací založenou na Corpus Pattern Analysis a lexikálním zdroji PDEV (Hanks a Pustejovsky, 2005) ukázaly potřebu evaluační míry, která by identifikovala optimální vztah mezi sémantickou granularitou sémantických kategorií v lexikálním popisu slovesa a spolehlivostí anotace, která se měří pomocí mezianotátorské shody. Představujeme takovou míru.
English abstract Experiments with semantic annotation based on the Corpus pattern Analysis and the lexical resource PDEV (Hanks and Pustejovsky, 2005), revealed a need of an evaluation measure that would identify the optimum relation between the semantic granularity of the semantic categories in the description of a verb and the reliability of the annotation expressed by the interannotator agreement (IAA). We have introduced the Reliable Information Gain (RG), which computes this relation for each tag selected by the annotators and relates it to the entry as a whole, suggesting merges of unreliable tags whenever it would increase the information gain of the entire tagset (the number of semantic categories in an entry). The merges suggested in our 19-verb sample correspond with common sense. One of the possible applications of this measure is quality management of the entries in a lexical resource.
Specialization linguistics ("jazykověda")
Confidentiality default – not confidential
Open access no
Editor(s)* Ruth Vatvedt Fjeld; Julie Matilde Torjusen
ISBN* 978-82-303-2228-4
Address* Oslo, Norway
Month* August
Institution* Department of Linguistics and Scandinavian Studies, University of Oslo
Creator: Common Account
Created: 10/16/12 10:39 AM
Modifier: Common Account
Modified: 11/23/15 11:20 AM
***

Optimizing semantic granularity for NLP - report ...publicpp523-531 Cinkova, Holub and Kriz.pdfapplication/pdf
Content, Design & Functionality: ÚFAL, 2006–2016. Page generated: Mon Nov 20 12:59:36 CET 2017

[ Back to the navigation ] [ Back to the content ]

100% OpenAIRE compliant