NER for Albanian Language: A Manually Annotated Corpus and Machine Learning Models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F26%3A34R253AY" target="_blank" >RIV/00216208:11320/26:34R253AY - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-3-031-87769-8_14" target="_blank" >http://dx.doi.org/10.1007/978-3-031-87769-8_14</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-87769-8_14" target="_blank" >10.1007/978-3-031-87769-8_14</a>

Jazyk výsledku

angličtina

Název v původním jazyce

NER for Albanian Language: A Manually Annotated Corpus and Machine Learning Models

Popis výsledku v původním jazyce

Recent advancements in artificial intelligence (AI) have significantly enhanced tasks like named entity recognition (NER), enabling the identification of people, organizations, products, events, places, and dates. This paper introduces an Albanian NER corpus with 1,003,836 tokens (56,595 sentences), including 89,850 labelled tokens, annotated with 10 NER tags. The corpus, sourced from well-known Albanian news platforms, are used to train and evaluate 10 models using algorithms such as Naïve Bayes, Logistic Regression, SVM, Random Forest, Gradient Boosting, Extreme Gradient Boosting, and Multi-Layer Perceptron variants. Among these, Extra Trees and Random Forest achieved the best results, with approximately 96% accuracy and a 95% F1 score. As the largest NER corpus in Albanian, this resource advances linguistic research and AI applications, enhancing NER tasks and advance natural language processing (NLP) developments for the Albanian language. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Název v anglickém jazyce

NER for Albanian Language: A Manually Annotated Corpus and Machine Learning Models

Popis výsledku anglicky

Recent advancements in artificial intelligence (AI) have significantly enhanced tasks like named entity recognition (NER), enabling the identification of people, organizations, products, events, places, and dates. This paper introduces an Albanian NER corpus with 1,003,836 tokens (56,595 sentences), including 89,850 labelled tokens, annotated with 10 NER tags. The corpus, sourced from well-known Albanian news platforms, are used to train and evaluate 10 models using algorithms such as Naïve Bayes, Logistic Regression, SVM, Random Forest, Gradient Boosting, Extreme Gradient Boosting, and Multi-Layer Perceptron variants. Among these, Extra Trees and Random Forest achieved the best results, with approximately 96% accuracy and a 95% F1 score. As the largest NER corpus in Albanian, this resource advances linguistic research and AI applications, enhancing NER tasks and advance natural language processing (NLP) developments for the Albanian language. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

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 periodika

Lecture Notes on Data Engineering and Communications Technologies

ISSN

23674512

e-ISSN

—

Svazek periodika

247

Číslo periodika v rámci svazku

2025

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

13

Strana od-do

153-165

Kód UT WoS článku

—

EID výsledku v databázi Scopus

2-s2.0-105003097414