Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
Using Few-Shot Learning Techniques for Named Entity Recognition and Relation Extraction. / Bondarenko, Ivan; Berezin, Sergey; Pauls, Alexey и др.
Proceedings - 2020 Science and Artificial Intelligence Conference, S.A.I.ence 2020. Novosibirsk, Russia : Institute of Electrical and Electronics Engineers Inc., 2020. стр. 58-65 9303192 (Proceedings - 2020 Science and Artificial Intelligence Conference, S.A.I.ence 2020).Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
}
TY - GEN
T1 - Using Few-Shot Learning Techniques for Named Entity Recognition and Relation Extraction
AU - Bondarenko, Ivan
AU - Berezin, Sergey
AU - Pauls, Alexey
AU - Batura, Tatiana
AU - Rubtsova, Yuliya
AU - Tuchinov, Bair
N1 - Publisher Copyright: © 2020 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2020/11/14
Y1 - 2020/11/14
N2 - This paper presents new methods for entity recognition and relation extraction tasks on partially labeled and unlabeled datasets. The proposed methods are based on techniques of semi-supervised, unsupervised and the transfer learning. We use the few-shot learning technique to construct specific algorithms for the new data sources without manual retraining. To compare the results with other studies, we conducted experiments on two benchmark datasets for the Russian language. The results for named entity recognition demonstrate significant improvement and outperform the state-of-the-art results. Our results for relation extraction are comparable to other research. We assume that a longer BERT fine-tuning will help to improve them, and we also plan to experiment with other few-shot learning methods in the near future.
AB - This paper presents new methods for entity recognition and relation extraction tasks on partially labeled and unlabeled datasets. The proposed methods are based on techniques of semi-supervised, unsupervised and the transfer learning. We use the few-shot learning technique to construct specific algorithms for the new data sources without manual retraining. To compare the results with other studies, we conducted experiments on two benchmark datasets for the Russian language. The results for named entity recognition demonstrate significant improvement and outperform the state-of-the-art results. Our results for relation extraction are comparable to other research. We assume that a longer BERT fine-tuning will help to improve them, and we also plan to experiment with other few-shot learning methods in the near future.
KW - few-shot learning
KW - Named entity recognition
KW - relation extraction
KW - transfer learning
UR - http://www.scopus.com/inward/record.url?scp=85099582329&partnerID=8YFLogxK
U2 - 10.1109/S.A.I.ence50533.2020.9303192
DO - 10.1109/S.A.I.ence50533.2020.9303192
M3 - Conference contribution
AN - SCOPUS:85099582329
SN - 978-0-7381-3112-2
T3 - Proceedings - 2020 Science and Artificial Intelligence Conference, S.A.I.ence 2020
SP - 58
EP - 65
BT - Proceedings - 2020 Science and Artificial Intelligence Conference, S.A.I.ence 2020
PB - Institute of Electrical and Electronics Engineers Inc.
CY - Novosibirsk, Russia
T2 - 2020 Science and Artificial Intelligence Conference, S.A.I.ence 2020
Y2 - 14 November 2020 through 15 November 2020
ER -
ID: 27526694