A review on spectral clustering and stochastic block models

Clustering is an important statistical tool for the analysis of unsupervised data. Spectral clustering and stochastic block models, based on networks and graphs, are well established and widely used for community detection among many clustering algorithms. In this paper we review and discuss importa...

Full description

Saved in:
Bibliographic Details
Published inJournal of the Korean Statistical Society Vol. 50; no. 3; pp. 818 - 831
Main Authors Baek, Mina, Kim, Choongrak
Format Journal Article
LanguageEnglish
Published Singapore Springer Singapore 01.09.2021
한국통계학회
Subjects
Applied Statistics
Bayesian Inference
Mathematics and Statistics
Review Article
Statistical Theory and Methods
Statistics
Statistics and Computing/Statistics Programs
통계학
Dimension reduction
Spectral embedding
Manifold
Regularization
Random graph
Online AccessGet full text

Cover

More Information
Summary:Clustering is an important statistical tool for the analysis of unsupervised data. Spectral clustering and stochastic block models, based on networks and graphs, are well established and widely used for community detection among many clustering algorithms. In this paper we review and discuss important statistical issues in spectral clustering and stochastic block models.
ISSN:1226-3192
2005-2863
DOI:10.1007/s42952-021-00112-w