Initialization-similarity clustering algorithm

Simple item page
dc.citation.issue23
dc.citation.volume78
dc.contributor.authorLiu T
dc.contributor.authorZhu J
dc.contributor.authorZhou J
dc.contributor.authorZhu Y
dc.contributor.authorZhu X
dc.date.available2019-12
dc.date.issued2019-12
dc.descriptionCAUL read and publish agreement 2022
dc.description.abstractClassic k-means clustering algorithm randomly selects centroids for initialization to possibly output unstable clustering results. Moreover, random initialization makes the clustering result hard to reproduce. Spectral clustering algorithm is a two-step strategy, which first generates a similarity matrix and then conducts eigenvalue decomposition on the Laplacian matrix of the similarity matrix to obtain the spectral representation. However, the goal of the first step in the spectral clustering algorithm does not guarantee the best clustering result. To address the above issues, this paper proposes an Initialization-Similarity (IS) algorithm which learns the similarity matrix and the new representation in a unified way and fixes initialization using the sum-of-norms regularization to make the clustering more robust. The experimental results on ten real-world benchmark datasets demonstrate that our IS clustering algorithm outperforms the comparison clustering algorithms in terms of three evaluation metrics for clustering algorithm including accuracy (ACC), normalized mutual information (NMI), and Purity.
dc.description.publication-statusPublished
dc.format.extent33279 - 33296
dc.identifierhttp://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000500000600032&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=c5bb3b2499afac691c2e3c1a83ef6fef
dc.identifier.citationMULTIMEDIA TOOLS AND APPLICATIONS, 2019, 78 (23), pp. 33279 - 33296
dc.identifier.doi10.1007/s11042-019-7663-8
dc.identifier.eissn1573-7721
dc.identifier.elements-id423521
dc.identifier.harvestedMassey_Dark
dc.identifier.issn1380-7501
dc.publisherSpringer Science+Business Media, LLC
dc.relation.isPartOfMULTIMEDIA TOOLS AND APPLICATIONS
dc.subjectk-means clustering
dc.subjectSpectral clustering
dc.subjectInitialization
dc.subjectSimilarity
dc.subject.anzsrc0803 Computer Software
dc.subject.anzsrc0805 Distributed Computing
dc.subject.anzsrc0806 Information Systems
dc.subject.anzsrc0801 Artificial Intelligence and Image Processing
dc.titleInitialization-similarity clustering algorithm
dc.typeJournal article
pubs.notesNot known
pubs.organisational-group/Massey University
pubs.organisational-group/Massey University/College of Sciences
pubs.organisational-group/Massey University/College of Sciences/School of Mathematical and Computational Sciences
pubs.organisational-group/Massey University/College of Sciences/School of Natural and Computational Sciences
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Liu2019_Article_Initialization-similarityClust.pdf
Size:
1.35 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Journal Articles