Show simple item record

Generating common-sense scene graphs using a knowledge base BERT model

dc.rights.licenseCC-BY-NC-ND
dc.contributor.advisorDeoskar, Tejaswini
dc.contributor.authorSlewe, Chris
dc.date.accessioned2021-12-01T00:00:19Z
dc.date.available2021-12-01T00:00:19Z
dc.date.issued2021
dc.identifier.urihttps://studenttheses.uu.nl/handle/20.500.12932/260
dc.description.sponsorshipUtrecht University
dc.language.isoEN
dc.subjectScene graphs can be used to improve upon autonomous robots by describing a variety of environments. This thesis compares performance of transformer models, trained on common knowledge bases such as Wikipedia and ConceptNet, in the creation of common-sense graphs.
dc.titleGenerating common-sense scene graphs using a knowledge base BERT model
dc.type.contentMaster Thesis
dc.rights.accessrightsOpen Access
dc.subject.keywordsNLP, Transformer models, BERT, scene graphs, knowledge bases
dc.subject.courseuuArtificial Intelligence
dc.thesis.id809


Files in this item

Thumbnail
Name:
Msc_thesis_chris_slewe.pdf
Size:
3.674Mb
Format:
PDF
View/Open

This item appears in the following Collection(s)

Show simple item record