Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/29495
DC FieldValueLanguage
dc.contributor.authorTsapatsoulis, Nicolas-
dc.date.accessioned2023-06-26T08:59:30Z-
dc.date.available2023-06-26T08:59:30Z-
dc.date.issued2022-09-07-
dc.identifier.citationSETN '22: Proceedings of the 12th Hellenic Conference on Artificial Intelligence, 7 - 9 September 2022, Corfu, Greece, pp 1–7en_US
dc.identifier.isbn9781450395977-
dc.identifier.urihttps://hdl.handle.net/20.500.14279/29495-
dc.description.abstractThe existence of pre-trained deep learning models for image classification, such as those trained on the well-known Resnet-50 architecture, allows for easy application of transfer learning to several domains including image retrieval. Recently, we proposed topic modelling for the retrieval of Instagram photos based on the associated hashtags. In this paper we compare content-based image classification, based on transfer learning, with the classification based on topic modelling of Instagram hashtags for a set of 24 different concepts. The comparison was performed on a set of 1944 Instagram photos, 81 per concept. Despite the excellent performance of the pre-trained deep learning models, it appears that text-based retrieval, as performed by the topic models of Instagram hashtags, stills perform better.en_US
dc.formatpdfen_US
dc.language.isoenen_US
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectimage classificationen_US
dc.subjecttransfer learningen_US
dc.subjecttopic modellingen_US
dc.subjectdeep learningen_US
dc.titleClassification of Instagram photos: Topic modelling vs transfer learningen_US
dc.typeConference Papersen_US
dc.collaborationCyprus University of Technologyen_US
dc.subject.categoryMedia and Communicationsen_US
dc.journalsOpen Accessen_US
dc.countryCyprusen_US
dc.subject.fieldSocial Sciencesen_US
dc.publicationPeer Revieweden_US
dc.identifier.doi10.1145/3549737.3549759en_US
dc.identifier.scopus2-s2.0-85138333325-
dc.identifier.urlhttps://api.elsevier.com/content/abstract/scopus_id/85138333325-
cut.common.academicyear2022-2023en_US
dc.identifier.spage1en_US
dc.identifier.epage7en_US
item.fulltextWith Fulltext-
item.openairecristypehttp://purl.org/coar/resource_type/c_c94f-
item.openairetypeconferenceObject-
item.grantfulltextopen-
item.languageiso639-1en-
item.cerifentitytypePublications-
crisitem.author.deptDepartment of Communication and Marketing-
crisitem.author.facultyFaculty of Communication and Media Studies-
crisitem.author.orcid0000-0002-6739-8602-
crisitem.author.parentorgFaculty of Communication and Media Studies-
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation
Files in This Item:
File Description SizeFormat
tsapatsoulis.pdffull text1.16 MBAdobe PDFView/Open
CORE Recommender
Show simple item record

Page view(s) 50

141
Last Week
2
Last month
3
checked on Feb 14, 2025

Download(s) 50

58
checked on Feb 14, 2025

Google ScholarTM

Check

Altmetric


This item is licensed under a Creative Commons License Creative Commons