In-depth evaluation of Romanian natural language processing pipelines

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A90101%2F21%3A10441616" target="_blank" >RIV/00216208:90101/21:10441616 - isvavai.cz</a>

Výsledek na webu

<a href="https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=4EPuFfhHFg" target="_blank" >https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=4EPuFfhHFg</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

In-depth evaluation of Romanian natural language processing pipelines

Popis výsledku v původním jazyce

With the increased size of Universal Dependencies tree banks, several basic language processing kits (BLARK) for multiple languages appeared in recent years, indicating improved performances on different languages. Nevertheless, published results are not directly comparable for the Romanian language since different tools make use of different Universal Dependencies versions and different additional resources, such as pre-trained word embeddings. In this paper, we re-train several state-of-the-art tools for processing Romanian language by using a common methodology comprising of training and evaluating on the same version of RoRefTrees corpus and using the same pre-trained word embeddings from the representative corpus of contemporary Romanian language (CoRoLa). Furthermore, we also explore the capabilities of the trained models when faced with unseen text from a different domain. For this purpose, we further test the resulting model on the SiMoNERo corpus. We employ different metrics to assess the performance on operations like tokenization, sentence splitting, lemmatization, part-of-speech tagging and dependency parsing.

Název v anglickém jazyce

In-depth evaluation of Romanian natural language processing pipelines

Popis výsledku anglicky

With the increased size of Universal Dependencies tree banks, several basic language processing kits (BLARK) for multiple languages appeared in recent years, indicating improved performances on different languages. Nevertheless, published results are not directly comparable for the Romanian language since different tools make use of different Universal Dependencies versions and different additional resources, such as pre-trained word embeddings. In this paper, we re-train several state-of-the-art tools for processing Romanian language by using a common methodology comprising of training and evaluating on the same version of RoRefTrees corpus and using the same pre-trained word embeddings from the representative corpus of contemporary Romanian language (CoRoLa). Furthermore, we also explore the capabilities of the trained models when faced with unseen text from a different domain. For this purpose, we further test the resulting model on the SiMoNERo corpus. We employ different metrics to assess the performance on operations like tokenization, sentence splitting, lemmatization, part-of-speech tagging and dependency parsing.

Druh

J_imp - Článek v periodiku v databázi Web of Science

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

2021

Kód důvěrnosti údajů

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

Název periodika

Romanian Journal of Information Science and Technology

ISSN

1453-8245

e-ISSN

—

Svazek periodika

24

Číslo periodika v rámci svazku

4

Stát vydavatele periodika

RO - Rumunsko

Počet stran výsledku

18

Strana od-do

384-401

Kód UT WoS článku

000731880700004

EID výsledku v databázi Scopus

—