Named Entity Discovery and Alignment in Parallel Data.

D 2025

Named Entity Discovery and Alignment in Parallel Data.

NEVĚŘILOVÁ, Zuzana

Základní údaje

Originální název

Named Entity Discovery and Alignment in Parallel Data.

Autoři

NEVĚŘILOVÁ, Zuzana

Vydání

Volume 3. Porto (Portugal), Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025), od s. 1215-1220, 6 s. 2025

Nakladatel

SCITEPRESS – Science and Technology Publications, Lda.

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

60200 6.2 Languages and Literature

Stát vydavatele

Portugalsko

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

URL

Označené pro přenos do RIV

Organizační jednotka

Fakulta informatiky

ISBN

978-989-758-737-5

ISSN

Klíčová slova anglicky

Named Entity Recognition; Named Entity Alignment; Named Entity Discovery; Named Entity Linking

Štítky

core_B, firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 27. 2. 2025 14:29, RNDr. Zuzana Nevěřilová, Ph.D.

Anotace

V originále

The paper describes two experiments with named entity discovery and alignment for English-Czech parallel data. In the previous work, we enriched the Parallel Global Voices corpus with named entity recognition (NER) for both languages and named entity linking (NEL) annotations for English. The alignment experiment employs sentence transformers and cosine similarity to identify NE translations from English to Czech and possibly other languages. The discovery experiment uses the same method to find possible translations between named entities in English and Czech n-grams. The described method achieves an F1 score of 0.94 in finding alignments between recognized entities. However, the same method can also discover unknown named entities with an F1 score of 0.70. The result indicates the method can be used to recognize named entities in parallel data in cases where no NER model is available with sufficient quality.

Návaznosti

LM2023062, projekt VaV

Název: Digitální výzkumná infrastruktura pro jazykové technologie, umění a humanitní vědy

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, LINDAT/CLARIAH-CZ - Digitální výzkumná infrastruktura pro jazykové technologie, umění a humanitní vědy

Citovat

NEVĚŘILOVÁ, Zuzana. Named Entity Discovery and Alignment in Parallel Data. Online. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025). Volume 3. Porto (Portugal): SCITEPRESS – Science and Technology Publications, Lda., 2025, s. 1215-1220. ISBN 978-989-758-737-5.

@inproceedings{2480339,
   author = {Nevěřilová, Zuzana},
   address = {Porto (Portugal)},
   booktitle = {Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025)},
   edition = {Volume 3},
   keywords = {Named Entity Recognition; Named Entity Alignment; Named Entity Discovery; Named Entity Linking},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Porto (Portugal)},
   isbn = {978-989-758-737-5},
   pages = {1215-1220},
   publisher = {SCITEPRESS – Science and Technology Publications, Lda.},
   title = {Named Entity Discovery and Alignment in Parallel Data.},
   url = {https://www.insticc.org/node/TechnicalProgram/ICAART/2025/presentationDetails/133113},
   year = {2025}
}

TY  - CONF
ID  - 2480339
AU  - Nevěřilová, Zuzana
PY  - 2025
TI  - Named Entity Discovery and Alignment in Parallel Data.
PB  - SCITEPRESS – Science and Technology Publications, Lda.
CY  - Porto (Portugal)
SN  - 9789897587375
KW  - Named Entity Recognition
KW  - Named Entity Alignment
KW  - Named Entity Discovery
KW  - Named Entity Linking
UR  - https://www.insticc.org/node/TechnicalProgram/ICAART/2025/presentationDetails/133113
N2  - The paper describes two experiments with named entity discovery and alignment for English-Czech parallel data. In the previous work, we enriched the Parallel Global Voices corpus with named entity recognition (NER) for both languages and named entity linking (NEL) annotations for English. The alignment experiment employs sentence transformers and cosine similarity to identify NE translations from English to Czech and possibly other languages. The discovery experiment uses the same method to find possible translations between named entities in English and Czech n-grams. The described method achieves an F1 score of 0.94 in finding alignments between recognized entities. However, the same method can also discover unknown named entities with an F1 score of 0.70. The result indicates the method can be used to recognize named entities in parallel data in cases where no NER model is available with sufficient quality.
ER  -

NEVĚŘILOVÁ, Zuzana. Named Entity Discovery and Alignment in Parallel Data. Online. In \textit{Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025)}. Volume 3. Porto (Portugal): SCITEPRESS – Science and Technology Publications, Lda., 2025, s.~1215-1220. ISBN~978-989-758-737-5.

Přehled o publikaci