Full metadata record
DC poleHodnotaJazyk
dc.contributor.authorOrtiz Pablo, Dalia
dc.contributor.authorBadri, Sushruth
dc.contributor.authorNorén, Erik
dc.contributor.authorNötzli, Christoph
dc.contributor.editorSkala, Václav
dc.date.accessioned2023-10-03T15:28:47Z
dc.date.available2023-10-03T15:28:47Z
dc.date.issued2023
dc.identifier.citationJournal of WSCG. 2023, vol. 31, no. 1-2, p. 53-62.en
dc.identifier.issn1213 – 6972 (hard copy)
dc.identifier.issn1213 – 6980 (CD-ROM)
dc.identifier.issn1213 – 6964 (on-line)
dc.identifier.urihttp://hdl.handle.net/11025/54284
dc.format10 s.cs
dc.format.mimetypeapplication/pdf
dc.language.isoenen
dc.publisherVáclav Skala - UNION Agencyen
dc.rights© Václav Skala - UNION Agencyen
dc.subjectklasifikace obrázkůcs
dc.subjectspravedlnostcs
dc.subjectzmírnění předsudkůcs
dc.subjectklasifikace pohlavícs
dc.subjectpřenos učenícs
dc.subjectsbírka lidského dědictvícs
dc.titleBias mitigation techniques in image classification: fair machine learning in human heritage collectionsen
dc.typečlánekcs
dc.typearticleen
dc.rights.accessopenAccessen
dc.type.versionpublishedVersionen
dc.description.abstract-translatedA major problem with using automated classification systems is that if they are not engineered correctly and with fairness considerations, they could be detrimental to certain populations. Furthermore, while engineers have developed cutting-edge technologies for image classification, there is still a gap in the application of these models in human heritage collections, where data sets usually consist of low-quality pictures of people with diverse ethnicity, gender, and age. In this work, we evaluate three bias mitigation techniques using two state-of-the-art neural networks, Xception and EfficientNet, for gender classification. Moreover, we explore the use of transfer learning using a fair data set to overcome the training data scarcity. We evaluated the effectiveness of the bias mitigation pipeline on a cultural heritage collection of photographs from the 19th and 20th centuries, and we used the FairFace data set for the transfer learning experiments. After the evaluation, we found that transfer learning is a good technique that allows better performance when working with a small data set. Moreover, the fairest classifier was found to be accomplished using transfer learning, threshold change, re-weighting and image augmentation as bias mitigation methodsen
dc.subject.translatedimage classificationen
dc.subject.translatedfairnessen
dc.subject.translatedbias mitigationen
dc.subject.translatedgender classificationen
dc.subject.translatedtransfer learningen
dc.subject.translatedHuman Heritage Collectionen
dc.identifier.doihttps://www.doi.org/10.24132/JWSCG.2023.6
dc.type.statusPeer-revieweden
Vyskytuje se v kolekcích:Volume 31, Number 1-2 (2023)

Soubory připojené k záznamu:
Soubor Popis VelikostFormát 
!_2023-Journal_WSCG-63-72.pdfPlný text1,23 MBAdobe PDFZobrazit/otevřít
Zobrazit minimální záznam Zobrazit statistiky


Použijte tento identifikátor k citaci nebo jako odkaz na tento záznam: http://hdl.handle.net/11025/54284

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.

hledání
navigace
  1. DSpace at University of West Bohemia
  2. Publikace ZČU / UWB Publications
  3. Časopisy ZČU / UWB Journals
  4. Journal of WSCG
  5. Volume 31 (2023)
  6. Volume 31, Number 1-2 (2023)