Less Labelled Learning

Simple item page
dc.contributor.advisorFarquhar, J.D.R.
dc.contributor.advisorRaaijmakers, S.
dc.contributor.advisorBouma, H.
dc.contributor.advisorBoer, M. de
dc.contributor.authorHagendoorn, L.K.
dc.contributor.otherTNOen_US
dc.date.issued2017-08-25
dc.description.abstractFor most of us humans, extracting information from visual data comes very natural. For a computer however, recognising what it is looking at is a challenge. To teach the computer to recognise the objects in an image, it needs to be trained on a large dataset with many annotated images. Creating these annotations is a time consuming and costly process, and the number of unlabelled images available will always greatly outnumber the number of labelled images. We propose a novel method for semi-supervised learning: Ordered ACOL-PL. We show that our method is able to achieve a competitive classification accuracy based on few samples in the training set accompanied by a class label. We also explore the dependencies of the method on the selected and learned feature space, as well as the dependency on the superclasses used.en_US
dc.identifier.urihttp://theses.ubn.ru.nl/handle/123456789/5234
dc.language.isoenen_US
dc.thesis.facultyFaculteit der Sociale Wetenschappenen_US
dc.thesis.specialisationMaster Artificial Intelligenceen_US
dc.thesis.studyprogrammeArtificial Intelligenceen_US
dc.thesis.typeMasteren_US
dc.titleLess Labelled Learningen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Hagendoorn, L._MSc_Thesis_2017.pdf
Size:
1.62 MB
Format:
Adobe Portable Document Format
Description:
Thesis text
Download
Collections
Faculteit der Sociale Wetenschappen