Opravidlo 2.0: AI-powered Proofreader for Czech Texts

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F26%3AEGS6QHRH" target="_blank" >RIV/00216208:11320/26:EGS6QHRH - isvavai.cz</a>

Výsledek na webu

<a href="https://research.brighton.ac.uk/files/60194662/CLARIN2025_ConferenceProceedings.pdf#page=132" target="_blank" >https://research.brighton.ac.uk/files/60194662/CLARIN2025_ConferenceProceedings.pdf#page=132</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Opravidlo 2.0: AI-powered Proofreader for Czech Texts

Popis výsledku v původním jazyce

Opravidlo 2.0 is an advanced AI-powered proofreader for Czech that combines rule-based linguistic analysis with machine learning techniques to improve the accuracy and usability of automatic grammar correction. Building upon the success of Beta Opravidlo, a rule-based tool developed in 2022, the new version addresses its limitations—particularly low recall—by integrating neural models capable of handling complex linguistic phenomena. The hybrid architecture ensures explainability and minimal correction. The system is deployed as public service and supported by institutional infrastructure, with plans to release correction rules and annotated datasets aligned with FAIR principles and GDPR compliance. Integration with CLARIN and LINDAT/CLARIAH ensures sustainability, interoperability, and future extensibility to other Slavic languages. Opravidlo 2.0 thus represents a significant step forward in the development of language technology for morphologically rich languages.

Název v anglickém jazyce

Opravidlo 2.0: AI-powered Proofreader for Czech Texts

Popis výsledku anglicky

Opravidlo 2.0 is an advanced AI-powered proofreader for Czech that combines rule-based linguistic analysis with machine learning techniques to improve the accuracy and usability of automatic grammar correction. Building upon the success of Beta Opravidlo, a rule-based tool developed in 2022, the new version addresses its limitations—particularly low recall—by integrating neural models capable of handling complex linguistic phenomena. The hybrid architecture ensures explainability and minimal correction. The system is deployed as public service and supported by institutional infrastructure, with plans to release correction rules and annotated datasets aligned with FAIR principles and GDPR compliance. Integration with CLARIN and LINDAT/CLARIAH ensures sustainability, interoperability, and future extensibility to other Slavic languages. Opravidlo 2.0 thus represents a significant step forward in the development of language technology for morphologically rich languages.

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

2025

Kód důvěrnosti údajů

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

Název statě ve sborníku

CLARIN Annual Conference Proceedings

ISBN

—

ISSN

2773-2177

e-ISSN

—

Počet stran výsledku

4

Strana od-do

123-126

Název nakladatele

—

Místo vydání

—

Místo konání akce

Vienna, Austria

Datum konání akce

1. 1. 2026

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—