ResearchSpace

Improving short text classification through global augmentation methods

Show simple item record

dc.contributor.author Marivate, V
dc.contributor.author Sefara, Tshephisho J
dc.date.accessioned 2020-09-14T18:41:27Z
dc.date.available 2020-09-14T18:41:27Z
dc.date.issued 2020-08
dc.identifier.citation Marivate, V. & Sefara, T.J. 2020. Improving short text classification through global augmentation methods. In International Cross-Domain Conference for Machine Learning and Knowledge Extraction. Springer, Cham, pp. 385-399 en_US
dc.identifier.isbn 978-3-030-57321-8
dc.identifier.isbn 978-3-030-57320-1
dc.identifier.uri DOI: https://doi.org/10.1007/978-3-030-57321-8_21
dc.identifier.uri https://link.springer.com/chapter/10.1007%2F978-3-030-57321-8_21
dc.identifier.uri http://hdl.handle.net/10204/11579
dc.description Copyright: 2020 Springer. Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, please consult the publisher's website. The definitive version of the work is published in Holzinger A., Kieseberg P., Tjoa A., Weippl E. (eds) Machine Learning and Knowledge Extraction. CD-MAKE 2020. Lecture Notes in Computer Science, vol 12279. Springer, Cham. https://doi.org/10.1007/978-3-030-57321-8_21 en_US
dc.description.abstract We study the effect of different approaches to text augmentation. To do this we use three datasets that include social media and formal text in the form of news articles. Our goal is to provide insights for practitioners and researchers on making choices for augmentation for classification use cases. We observe that Word2Vec-based augmentation is a viable option when one does not have access to a formal synonym model (like WordNet-based augmentation). The use of mixup further improves performance of all text based augmentations and reduces the effects of overfitting on a tested deep learning model. Round-trip translation with a translation service proves to be harder to use due to cost and as such is less accessible for both normal and low resource use-cases. en_US
dc.language.iso en en_US
dc.relation.ispartofseries Workflow;23683
dc.subject Data augmentation en_US
dc.subject Deep neural networks en_US
dc.subject Natural Language Processing en_US
dc.subject Text classification en_US
dc.title Improving short text classification through global augmentation methods en_US
dc.type Book Chapter en_US
dc.identifier.apacitation Marivate, V., & Sefara, T. J. (2020). Improving short text classification through global augmentation methods., <i>Workflow;23683</i> http://hdl.handle.net/10204/11579 en_ZA
dc.identifier.chicagocitation Marivate, V, and Tshephisho J Sefara. "Improving short text classification through global augmentation methods" In <i>WORKFLOW;23683</i>, n.p.: n.p. 2020. http://hdl.handle.net/10204/11579. en_ZA
dc.identifier.vancouvercitation Marivate V, Sefara TJ. Improving short text classification through global augmentation methods.. Workflow;23683. [place unknown]: [publisher unknown]; 2020. [cited yyyy month dd]. http://hdl.handle.net/10204/11579. en_ZA
dc.identifier.ris TY - Book Chapter AU - Marivate, V AU - Sefara, Tshephisho J AB - We study the effect of different approaches to text augmentation. To do this we use three datasets that include social media and formal text in the form of news articles. Our goal is to provide insights for practitioners and researchers on making choices for augmentation for classification use cases. We observe that Word2Vec-based augmentation is a viable option when one does not have access to a formal synonym model (like WordNet-based augmentation). The use of mixup further improves performance of all text based augmentations and reduces the effects of overfitting on a tested deep learning model. Round-trip translation with a translation service proves to be harder to use due to cost and as such is less accessible for both normal and low resource use-cases. DA - 2020-08 DB - ResearchSpace DP - CSIR KW - Data augmentation KW - Deep neural networks KW - Natural Language Processing KW - Text classification LK - https://researchspace.csir.co.za PY - 2020 SM - 978-3-030-57321-8 SM - 978-3-030-57320-1 T1 - Improving short text classification through global augmentation methods TI - Improving short text classification through global augmentation methods UR - http://hdl.handle.net/10204/11579 ER - en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record