A multimodal deep learning architecture for smoking detection with a small data approach

A tétel rövid nézete
dc.contributor.authorLakatos, Róbert
dc.contributor.authorPollner, Péter
dc.contributor.authorHajdu, András
dc.contributor.authorJoó, Tamás
dc.contributor.authorLakatos Róbert (1986-) (Informatikus)
dc.contributor.submitterdepAdattudomány és Vizualizáció Tanszék -- 905
dc.contributor.submitterdepIK
dc.contributor.submitterdepDebreceni Egyetem
dc.date.accessioned2024-02-29T12:15:06Z
dc.date.available2024-02-29T12:15:06Z
dc.date.oa2024-11-14
dc.date.updated2024-02-29T12:15:06Z
dc.description.abstractCovert tobacco advertisements often raise regulatory measures. This paper presents that artificial intelligence, particularly deep learning, has great potential for detecting hidden advertising and allows unbiased, reproducible, and fair quantification of tobacco-related media content. We propose an integrated text and image processing model based on deep learning, generative methods, and human reinforcement, which can detect smoking cases in both textual and visual formats, even with little available training data. Our model can achieve 74% accuracy for images and 98% for text. Furthermore, our system integrates the possibility of expert intervention in the form of human reinforcement. Using the pre-trained multimodal, image, and text processing models available through deep learning makes it possible to detect smoking in different media even with few training data.
dc.description.correctorLB
dc.identifier.citationFrontiers in Artificial Intelligence. -7 (2024), p. 1-8. -Front. Artif. Intell. -2624-8212
dc.identifier.doi10.3389/frai.2024.1326050
dc.identifier.issn2624-8212
dc.identifier.opachttps://ebib.lib.unideb.hu/ebib/CorvinaWeb?action=cclfind&resultview=long&ccltext=idno+BIBFORM118939
dc.identifier.urihttps://hdl.handle.net/2437/366932
dc.identifier.urlhttps://www.frontiersin.org/articles/10.3389/frai.2024.1326050/full
dc.languageeng
dc.rights.accessopen access journal
dc.subject.otheridegen nyelvű folyóiratközlemény külföldi lapban
dc.subject.otherAI supported preventive healthcare
dc.subject.otherpre-training with generative AI
dc.subject.othermultimodal deep learning
dc.subject.otherautomated assessment of covert advertisement
dc.subject.otherfew-shot learning
dc.subject.othersmoking detections
dc.tenderGINOP-2.3.2-15-2016-00005 -- GINOP
dc.tenderTKP2021-NKTA-34 -- Egyéb
dc.tenderKDP-2021 -- Egyéb
dc.tenderRRF-2.3.1-21-2022-00006 -- Egyéb
dc.titleA multimodal deep learning architecture for smoking detection with a small data approach
Fájlok
Eredeti köteg (ORIGINAL bundle)
Megjelenítve 1 - 1 (Összesen 1)
Nincs kép
Név:
FILE_UP_0_frai-07-1326050.pdf
Méret:
1.04 MB
Formátum:
Adobe Portable Document Format
Leírás:
Kiadói változat
Letöltés
Gyűjtemények
Publikációs Adatbázis