DC FieldValueLanguage
dc.contributor.authorNtalianis, Klimis-
dc.contributor.authorGiannoulakis, Stamatios-
dc.contributor.authorTsapatsoulis, Nicolas-
dc.date.accessioned2024-10-29T14:22:27Z-
dc.date.available2024-10-29T14:22:27Z-
dc.date.issued2018-03-29-
dc.identifierscopus-85038015389-
dc.identifier.isbn978-1-5386-1956-8-
dc.identifier.other85038015389-
dc.identifier.urihttps://uniwacris.uniwa.gr/handle/3000/2850-
dc.description.abstractMany automatic image annotation methods are based on the learning by example paradigm. Image tagging, through manual image inspection, is the first step towards this end. However, manual image annotation, even for creating the training sets, is time-consuming, complicated and contains human subjectivity errors. Thus, alternative ways for automatically creating training examples, i.e., pairs of images and tags, are crucial. As we showed in one of our previous studies, tags accompanying photos in social media and especially the Instagram hashtags can be used for image annotation. However, it turned out that only a 20% of the Instagram hashtags are actually relevant to the content of the image they accompany. Identifying those hashtags through crowdsourcing is a plausible solution. In this work, we investigate the effectiveness of the HITS algorithm for identifying the right tags in a crowdsourced image tagging scenario. For this purpose, we create a bipartite graph in which the first type of nodes corresponds to the annotators and the second type to the tags they select, among the hashtags, to annotate a particular Instagram image. From the results, we conclude that the authority value of the HITS algorithm provides an accurate estimation of the appropriateness of each Instagram hashtag to be used as a tag for the image it accompanies while the hub value can be used to filter out the dishonest annotators.en_US
dc.publisherIEEEen_US
dc.relation.ispartofProceedings of the 2017 IEEE 15th Intl Conf on Dependable, Autonomic and Secure Computing, 15th Intl Conf on Pervasive Intelligence and Computing, 3rd Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress(DASC/PiCom/DataCom/CyberSciTech)en_US
dc.subjectBipartite networken_US
dc.subjectGraphsen_US
dc.subjectHashtagsen_US
dc.subjectHITS algorithmen_US
dc.subjectImage retrievalen_US
dc.subjectImage taggingen_US
dc.subjectInstagramen_US
dc.subjectMachine learningen_US
dc.titleIdentifying image tags from instagram hashtags using the HITS algorithmen_US
dc.typeConference Paperen_US
dc.relation.conference2017 IEEE 15th Intl Conf on Dependable, Autonomic and Secure Computing, 15th Intl Conf on Pervasive Intelligence and Computing, 3rd Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress(DASC/PiCom/DataCom/CyberSciTech), 06-10 November 2017, Orlando, FL, USAen_US
dc.identifier.doi10.1109/DASC-PICom-DataCom-CyberSciTec.2017.29en_US
dc.identifier.scopus2-s2.0-85038015389-
dcterms.accessRights0en_US
dc.relation.deptDepartment of Business Administrationen_US
dc.relation.facultySchool of Administrative, Economics and Social Sciencesen_US
dc.identifier.spage89en_US
dc.identifier.epage94en_US
dc.collaborationUniversity of West Attica (UNIWA)en_US
dc.journalsOpen Accessen_US
dc.publicationPeer Revieweden_US
dc.countryGreeceen_US
local.metadatastatusnot verifieden_US
item.cerifentitytypePublications-
item.openairetypeConference Paper-
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.fulltextNo Fulltext-
item.grantfulltextnone-
crisitem.author.deptDepartment of Business Administration-
crisitem.author.facultySchool of Administrative, Economics and Social Sciences-
crisitem.author.parentorgSchool of Administrative, Economics and Social Sciences-
Appears in Collections:Book Chapter / Κεφάλαιο Βιβλίου
CORE Recommender
Show simple item record

SCOPUSTM   
Citations

12
checked on Oct 30, 2024

Page view(s)

5
checked on Nov 5, 2024

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.