A Novel Approach for Root Selection in the Dependency Parsing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3A47YD5RBN" target="_blank" >RIV/00216208:11320/25:47YD5RBN - isvavai.cz</a>

Výsledek na webu

<a href="https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195383928&partnerID=40&md5=1ea8705547b735a48e5c68f72605291f" target="_blank" >https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195383928&partnerID=40&md5=1ea8705547b735a48e5c68f72605291f</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

A Novel Approach for Root Selection in the Dependency Parsing

Popis výsledku v původním jazyce

Although syntactic analysis using the sequence labeling method is promising, it can be problematic when the labels sequence does not contain a root label. This can result in errors in the final parse tree when the postprocessing method assumes the first word as the root. In this paper, we present a novel postprocessing method for BERT-based dependency parsing as sequence labeling. Our method leverages the root's part of speech tag to select a more suitable root for the dependency tree, instead of using the default first token. We conducted experiments on nine dependency treebanks from different languages and domains, and demonstrated that our technique consistently improves the labeled attachment score (LAS) on most of them. © 2024 ELRA Language Resource Association.

Název v anglickém jazyce

A Novel Approach for Root Selection in the Dependency Parsing

Popis výsledku anglicky

Although syntactic analysis using the sequence labeling method is promising, it can be problematic when the labels sequence does not contain a root label. This can result in errors in the final parse tree when the postprocessing method assumes the first word as the root. In this paper, we present a novel postprocessing method for BERT-based dependency parsing as sequence labeling. Our method leverages the root's part of speech tag to select a more suitable root for the dependency tree, instead of using the default first token. We conducted experiments on nine dependency treebanks from different languages and domains, and demonstrated that our technique consistently improves the labeled attachment score (LAS) on most of them. © 2024 ELRA Language Resource Association.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

—

Návaznosti

—

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Název statě ve sborníku

Workshop Open-Source Arab. Corpora Process. Tools, OSACT Shar. Tasks Arab. LLMs Hallucination Dialect MSA Mach. Transl. LREC-COLING - Workshop Proc.

ISBN

978-249381436-4

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

46-49

Název nakladatele

European Language Resources Association (ELRA)

Místo vydání

—

Místo konání akce

Torino, Italia

Datum konání akce

1. 1. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—