Use this url to cite publication: https://hdl.handle.net/20.500.12259/145295
Testing performance of NER models for Russian
Type of publication
Straipsnis kitoje duomenų bazėje / Article in other database (S4)
Title [en]
Testing performance of NER models for Russian
Is part of
IJDATICS: International journal of design, analysis and tools for integrated circuits and systems. Hong Kong: Solari (HK) Co, 2021, Vol. 10, iss. 1
Date Issued
Date |
---|
2021 |
Publisher
Hong Kong: Solari (HK) Co
Is Referenced by
Extent
p. 55-58
Abstract (en)
This paper describes an experiment in testing 3 NER models (spaCy, Stanza and DeepPavlov) for Russian. The models were tested on WikiAnn Russian subset. Standard evaluation metrics (Precision, Recall and F-score) were complemented with 2 inter-annotator agreement measures (Fleiss’ κ and Krippendorff’s α). DeepPavlov performed better than the other 2 models in terms of all 3 standard performance measures. Stanza performed better than spaCy in terms of precision, while they shared the same Fmeasure score. Both Fleiss’ κ and Krippendorff’s α resulted in an only moderate agreement, showing, that our selected 3 NER models only moderately agree in terms of annotation of Russian named entities.
Type of document
type::text::journal::journal article::research article
Language
Anglų / English (en)
Coverage Spatial
Kinija / China (CN)
ISSN (of the container)
2223-523X
Other Identifier(s)
VDU02-000068391
Project(s)
GRACE, 883341
Access Rights
Atviroji prieiga / Open Access