A graph-based semi-supervised k nearest-neighbor method for nonlinear manifold distributed data classification

Simple item page
aut.relation.endpage688
aut.relation.startpage673
aut.relation.volume367-368en_NZ
aut.researcherKasabov, Nikola
dc.contributor.authorTu, Een_NZ
dc.contributor.authorZhang, Yen_NZ
dc.contributor.authorZhu, Len_NZ
dc.contributor.authorYang, Jen_NZ
dc.contributor.authorKasabov, Nen_NZ
dc.date.accessioned2016-08-10T04:32:48Z
dc.date.available2016-08-10T04:32:48Z
dc.date.copyright2016-11-01en_NZ
dc.date.issued2016-11-01en_NZ
dc.description.abstractk nearest neighbors (kNN) is one of the most widely used supervised learning algorithms to classify Gaussian distributed data, but it does not achieve good results when it is applied to nonlinear manifold distributed data, especially when a very limited amount of labeled samples are available. In this paper, we propose a new graph-based kNN algorithm which can effectively handle both Gaussian distributed data and nonlinear manifold distributed data. To achieve this goal, we first propose a constrained Tired Random Walk (TRW) by constructing an R-level nearest-neighbor strengthened tree over the graph, and then compute a TRW matrix for similarity measurement purposes. After this, the nearest neighbors are identified according to the TRW matrix and the class label of a query point is determined by the sum of all the TRW weights of its nearest neighbors. To deal with online situations, we also propose a new algorithm to handle sequential samples based a local neighborhood reconstruction. Comparison experiments are conducted on both synthetic data sets and real-world data sets to demonstrate the validity of the proposed new kNN algorithm and its improvements to other version of kNN algorithms. Given the widespread appearance of manifold structures in real-world problems and the popularity of the traditional kNN algorithm, the proposed manifold version kNN shows promising potential for classifying manifold-distributed data.en_NZ
dc.identifier.citationInformation Sciences, vol.367-368, pp.673 - 688en_NZ
dc.identifier.doi10.1016/j.ins.2016.07.016en_NZ
dc.identifier.issn0020-0255en_NZ
dc.identifier.urihttps://hdl.handle.net/10292/9986
dc.languageengen_NZ
dc.publisherElsevier Inc.en_NZ
dc.relation.urihttp://dx.doi.org/10.1016/j.ins.2016.07.016
dc.rightsCopyright © 2016 Elsevier Ltd. All rights reserved. This is the author’s version of a work that was accepted for publication in (see Citation). Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. The definitive version was published in (see Citation). The original publication is available at (see Publisher's Version).
dc.rights.accessrightsOpenAccessen_NZ
dc.subjectConstrained tired random walken_NZ
dc.subjectk Nearest neighborsen_NZ
dc.subjectManifold classificationen_NZ
dc.subjectSemi-Supervised learningen_NZ
dc.titleA graph-based semi-supervised k nearest-neighbor method for nonlinear manifold distributed data classificationen_NZ
dc.typeJournal Article
pubs.elements-id206102
pubs.organisational-data/AUT
pubs.organisational-data/AUT/Design & Creative Technologies
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
1606.00985v1.pdf
Size:
1.35 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
RE4.10 Grant of Licence.docx
Size:
14.05 KB
Format:
Microsoft Word 2007+
Description:
Download
Collections
KEDRI - the Knowledge Engineering and Discovery Research Institute