CouRGe: Counterfactual reviews generator for sentiment analysis

dc.contributor.authorCarraro, Diegoen
dc.contributor.authorBrown, Kenneth N.en
dc.contributor.funderScience Foundation Irelanden
dc.contributor.funderEuropean Regional Development Funden
dc.date.accessioned2023-03-29T12:05:27Z
dc.date.available2023-03-29T12:05:27Z
dc.date.issued2022-02-23en
dc.description.abstractPast literature in Natural Language Processing (NLP) has demonstrated that counterfactual data points are useful, for example, for increasing model generalisation, enhancing model interpretability, and as a data augmentation approach. However, obtaining counterfactual examples often requires human annotation effort, which is an expensive and highly skilled process. For these reasons, solutions that resort to transformer-based language models have been recently proposed to generate counterfactuals automatically, but such solutions show limitations. In this paper, we present CouRGe, a language model that, given a movie review (i.e. a seed review) and its sentiment label, generates a counterfactual review that is close (similar) to the seed review but of the opposite sentiment. CouRGe is trained by supervised fine-tuning of GPT-2 on a task-specific dataset of paired movie reviews, and its generation is prompt-based. The model does not require any modification to the network’s architecture or the design of a specific new task for fine-tuning. Experiments show that CouRGe’s generation is effective at flipping the seed sentiment and produces counterfactuals reasonably close to the seed review. This proves once again the great flexibility of language models towards downstream tasks as hard as counterfactual reasoning and opens up the use of CouRGe’s generated counterfactuals for the applications mentioned above.en
dc.description.statusPeer revieweden
dc.description.versionPublished Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationCarraro, D. and Brown, K. N. (2023) ‘Courge: counterfactual reviews generator for sentiment analysis’, AICS2022, in L. Longo and R. O’Reilly (eds) Artificial Intelligence and Cognitive Science. Cham: Springer Nature Switzerland, pp. 305–317. doi: 10.1007/978-3-031-26438-2_24.en
dc.identifier.doi10.1007/978-3-031-26438-2_24en
dc.identifier.endpage317en
dc.identifier.isbn978-3-031-26437-5en
dc.identifier.isbn978-3-031-26438-2en
dc.identifier.issued305en
dc.identifier.journaltitleCommunications in Computer and Information Scienceen
dc.identifier.startpage305en
dc.identifier.urihttps://hdl.handle.net/10468/14334
dc.identifier.volume1662en
dc.language.isoenen
dc.publisherSpringeren
dc.relation.ispartof30th Irish Conference on Artificial Intelligence and Cognitive Science (AICS2022), Munster Technological University, Cork, 8-9 Decemberen
dc.relation.projectinfo:eu-repo/grantAgreement/SFI/SFI Research Centres Programme::Phase 2/12/RC/2289-P2s/IE/INSIGHT Phase 2/en
dc.relation.projectinfo:eu-repo/grantAgreement/SFI/SFI Research Centres/12/RC/2289/IE/INSIGHT - Irelands Big Data and Analytics Research Centre/en
dc.rights© 2023 The Author(s). Open Access. This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were madeen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en
dc.subjectNatural language processingen
dc.subjectSentiment analysisen
dc.subjectLanguage modelsen
dc.subjectCounterfactual reasoningen
dc.subjectData augmentationen
dc.titleCouRGe: Counterfactual reviews generator for sentiment analysisen
dc.typeConference itemen
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
978-3-031-26438-2_24.pdf
Size:
335.98 KB
Format:
Adobe Portable Document Format
Description:
Published version
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
2.71 KB
Format:
Item-specific license agreed upon to submission
Description: