From web to dialects: how to enhance non-standard Russian lects lemmatisation?

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3AUBLWDWCL" target="_blank" >RIV/00216208:11320/23:UBLWDWCL - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2023.clasp-1.17/" target="_blank" >https://aclanthology.org/2023.clasp-1.17/</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

From web to dialects: how to enhance non-standard Russian lects lemmatisation?

Popis výsledku v původním jazyce

"The growing need for using small data distinguished by a set of distributional properties becomes all the more apparent in the era of large language models (LLM). In this paper, we show that for the lemmatisation of the web as corpora texts, heterogeneous social media texts, and dialect texts, the morphological tagging by a model trained on a small dataset with specific properties generally works better than the morphological tagging by a model trained on a large dataset. The material we use is Russian non-standard texts and interviews with dialect speakers. The sequence-to-sequence lemmatisation with the help of taggers trained on smaller linguistically aware datasets achieves the average results of 85 to 90 per cent. These results are consistently (but not always), by 1-2 per cent. higher than the results of lemmatisation with the help of the large-dataset-trained taggers. We analyse these results and outline the possible further research directions."

Název v anglickém jazyce

From web to dialects: how to enhance non-standard Russian lects lemmatisation?

Popis výsledku anglicky

"The growing need for using small data distinguished by a set of distributional properties becomes all the more apparent in the era of large language models (LLM). In this paper, we show that for the lemmatisation of the web as corpora texts, heterogeneous social media texts, and dialect texts, the morphological tagging by a model trained on a small dataset with specific properties generally works better than the morphological tagging by a model trained on a large dataset. The material we use is Russian non-standard texts and interviews with dialect speakers. The sequence-to-sequence lemmatisation with the help of taggers trained on smaller linguistically aware datasets achieves the average results of 85 to 90 per cent. These results are consistently (but not always), by 1-2 per cent. higher than the results of lemmatisation with the help of the large-dataset-trained taggers. We analyse these results and outline the possible further research directions."

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

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

"Proceedings of the 2023 CLASP Conference on Learning with Small Data (LSD)"

ISBN

979-8-89176-000-4

ISSN

2002-9764

e-ISSN

—

Počet stran výsledku

9

Strana od-do

167-175

Název nakladatele

Association for Computational Linguistics

Místo vydání

Gothenburg, Sweden

Místo konání akce

Gothenburg, Sweden

Datum konání akce

1. 1. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—