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

Socpus ID

85016397885 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85016397885

This document is currently not available here.

Share

COinS