Estimating POS Annotation Consistency of Different Treebanks in a Language

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10424428" target="_blank" >RIV/00216208:11320/20:10424428 - isvavai.cz</a>

Result on the web

<a href="https://www.aclweb.org/anthology/2020.tlt-1.9.pdf" target="_blank" >https://www.aclweb.org/anthology/2020.tlt-1.9.pdf</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Estimating POS Annotation Consistency of Different Treebanks in a Language

Original language description

We introduce a new symmetric metric (called θpos) that utilises the non-symmetric KLcpos3 metric (Rosa and Žabokrtský, 2015) to allow us to compare the annotation consistency between different annotated treebanks of a given language, when annotated under the same guidelines. We can set a threshold for this new metric so that a pair of treebanks can be considered harmonious in their annotation consistency if θpos surpasses the threshold. For the calculation of the threshold, we estimate the effects of (i) the size variation, and (ii) the genre variation in the considered pair of treebanks. The estimations are based on data from treebanks of distinct language families, making the threshold less dependent on the properties of individual languages. We demonstrate the utility of the proposed metric by listing the treebanks in Universal Dependencies version 2.5 (UDv2.5) (Zeman et al., 2019) data that are annotated consistently with other treebanks of the same language. However, the metric could be used to a

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2020

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 19th International Workshop on Treebanks and Linguistic Theories

ISBN

978-1-952148-01-9

ISSN

—

e-ISSN

—

Number of pages

18

Pages from-to

93-110

Publisher name

Heinrich-Heine-Universität Düsseldorf

Place of publication

Düsseldorf, Germany

Event location

Online

Event date

Oct 27, 2020

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—