TEAM-Atreides at SemEval-2022 Task 11: On leveraging data augmentation and ensemble to recognize complex Named Entities in Bangla
AffiliationSchool of Information, University of Arizona
MetadataShow full item record
CitationNazia Tasnim, Md. Istiak Shihab, Asif Shahriyar Sushmit, Steven Bethard, and Farig Sadeque. 2022. TEAM-Atreides at SemEval-2022 Task 11: On leveraging data augmentation and ensemble to recognize complex Named Entities in Bangla. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), pages 1524–1530, Seattle, United States. Association for Computational Linguistics.
JournalSemEval 2022 - 16th International Workshop on Semantic Evaluation, Proceedings of the Workshop
RightsCopyright © 2022 Association for Computational Linguistics. This is an open access article licensed on a Creative Commons Attribution 4.0 International License.
Collection InformationThis item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at firstname.lastname@example.org.
AbstractBiological and healthcare domains, artistic works, and organization names can all have nested, overlapping, discontinuous entity mentions that may be syntactically or semantically ambiguous in practice. Traditional sequence tagging algorithms are unable to recognize these complex mentions because they violate the assumptions upon which sequence tagging schemes are founded. In this paper, we describe our contribution to SemEval 2022 Task 11 on identifying such complex named entities. We leveraged an ensemble of ELECTRA-based models exclusively pretrained on the Bangla language with ELECTRA-based monolingual models pretrained on English to achieve competitive performance. Besides providing a system description, we also present the outcomes of our experiments on architectural decisions, dataset augmentations and post-competition findings. © 2022 Association for Computational Linguistics.
NoteOpen access journal
VersionFinal published version
Except where otherwise noted, this item's license is described as Copyright © 2022 Association for Computational Linguistics. This is an open access article licensed on a Creative Commons Attribution 4.0 International License.