Spectral Clustering Based On Local Pca
Keywords
Intersecting clusters; Local principal component analysis; Multi-manifold clustering; Spectral clustering
Abstract
We propose a spectral clustering method based on local principal components analysis (PCA). After performing local PCA in selected neighborhoods, the algorithm builds a nearest neighbor graph weighted according to a discrepancy between the principal subspaces in the neighborhoods, and then applies spectral clustering. As opposed to standard spectral methods based solely on pairwise distances between points, our algorithm is able to resolve intersections. We establish theoretical guarantees for simpler variants within a prototypical mathematical framework for multi-manifold clustering, and evaluate our algorithm on various simulated data sets.
Publication Date
3-1-2017
Publication Title
Journal of Machine Learning Research
Volume
18
Number of Pages
1-57
Document Type
Article
Personal Identifier
scopus
Copyright Status
Unknown
Socpus ID
85016397885 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85016397885
STARS Citation
Arias-Castro, Ery; Lerman, Gilad; and Zhang, Teng, "Spectral Clustering Based On Local Pca" (2017). Scopus Export 2015-2019. 4723.
https://stars.library.ucf.edu/scopus2015/4723