What can we learn from natural and artificial dependency trees

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F19%3A10427159" target="_blank" >RIV/00216208:11320/19:10427159 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.aclweb.org/anthology/W19-7915" target="_blank" >https://www.aclweb.org/anthology/W19-7915</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

What can we learn from natural and artificial dependency trees

Popis výsledku v původním jazyce

This paper is centered around two main contributions : the first one consists in introducing severalprocedures for generating random dependency trees with constraints; we later use these artificial treesto compare their properties with the properties of natural trees (i.e trees extracted from treebanks)and analyze the relationships between these properties in natural and artificial settings in order to findout which relationships are formally constrained and which are linguistically motivated. We take intoconsideration five metrics: tree length, height, maximum arity, mean dependency distance and meanflux weight, and also look into the distribution of local configurations of nodes. This analysis is basedon UD treebanks (version 2.3, Nivre et al. 2018) for four languages: Chinese, English, French and Japanese.

Název v anglickém jazyce

What can we learn from natural and artificial dependency trees

Popis výsledku anglicky

This paper is centered around two main contributions : the first one consists in introducing severalprocedures for generating random dependency trees with constraints; we later use these artificial treesto compare their properties with the properties of natural trees (i.e trees extracted from treebanks)and analyze the relationships between these properties in natural and artificial settings in order to findout which relationships are formally constrained and which are linguistically motivated. We take intoconsideration five metrics: tree length, height, maximum arity, mean dependency distance and meanflux weight, and also look into the distribution of local configurations of nodes. This analysis is basedon UD treebanks (version 2.3, Nivre et al. 2018) for four languages: Chinese, English, French and Japanese.

Druh

O - Ostatní výsledky

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

2019

Kód důvěrnosti údajů

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