Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/162087
Full metadata record
DC FieldValueLanguage
dc.contributor.authorFeng, Zijianen_US
dc.contributor.authorZhou, Hanzhangen_US
dc.contributor.authorZhu, Zixiaoen_US
dc.contributor.authorMao, Kezhien_US
dc.date.accessioned2022-10-04T02:29:45Z-
dc.date.available2022-10-04T02:29:45Z-
dc.date.issued2022-
dc.identifier.citationFeng, Z., Zhou, H., Zhu, Z. & Mao, K. (2022). Tailored text augmentation for sentiment analysis. Expert Systems With Applications, 205, 117605-. https://dx.doi.org/10.1016/j.eswa.2022.117605en_US
dc.identifier.issn0957-4174en_US
dc.identifier.urihttps://hdl.handle.net/10356/162087-
dc.description.abstractIn synonym replacement-based data augmentation techniques for natural language processing tasks, words in a sentence are often sampled randomly with equal probability. In this paper, we propose a novel data augmentation technique named Tailored Text Argumentation (TTA) for sentiment analysis. It has two main operations. The first operation is the probabilistic word sampling for synonym replacement based on the discriminative power and relevance of the word to sentiment. The second operation is the identification of words irrelevant to sentiment but discriminative for the training data, and application of zero masking or contextual replacement to these words. The first operation expands the coverage of discriminative words, while the second operation alleviates the problem of misfitting. Both operations tend to improve the model's generalization capability. Extensive experiments on simulated low-data regimes demonstrate that TTA yields notable improvements over six strong baselines. Finally, TTA is applied to public sentiment analysis on measures against Covid-19, which again proves the effectiveness of the new data augmentation algorithm.en_US
dc.description.sponsorshipNational Research Foundation (NRF)en_US
dc.language.isoenen_US
dc.relation.ispartofExpert Systems with Applicationsen_US
dc.rights© 2022 Elsevier Ltd. All rights reserved.en_US
dc.subjectEngineering::Electrical and electronic engineeringen_US
dc.titleTailored text augmentation for sentiment analysisen_US
dc.typeJournal Articleen
dc.contributor.schoolSchool of Electrical and Electronic Engineeringen_US
dc.contributor.schoolInterdisciplinary Graduate School (IGS)en_US
dc.identifier.doi10.1016/j.eswa.2022.117605-
dc.identifier.scopus2-s2.0-85131964557-
dc.identifier.volume205en_US
dc.identifier.spage117605en_US
dc.subject.keywordsSentiment Analysisen_US
dc.subject.keywordsText Augmentationen_US
dc.description.acknowledgementThis work is an outcome of the Future Resilient Systems project at Singapore-ETH Centre (SEC) supported by the National Research Foundation, Prime Minister’s Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme.en_US
item.grantfulltextnone-
item.fulltextNo Fulltext-
Appears in Collections:EEE Journal Articles
IGS Journal Articles

SCOPUSTM   
Citations 50

2
Updated on Feb 1, 2023

Page view(s)

44
Updated on Feb 2, 2023

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.