Year: 2020
Type: preprint
Abstract: Based on a Gaussian mixture type model of K components, we derive eigen selection procedures that improve the usual spectral clustering algorithms in high-dimensional settings, which typically act on the top few eigenvectors of an affinity matrix (e.g., X⊤X) derived from the data matrix X. Our selection principle formalizes two intuitions: (1) eigenvectors should be dropped when they have no clustering power; (2) some eigenvectors corresponding to smaller spiked eigenvalues should be dropped due to estimation inaccuracy. Our selection procedures lead to new spectral clustering algorithms: ESSC for K = 2 and GESSC for K > 2. The newly proposed algorithms enjoy better stability and compare favorably against canonical alternatives, as demonstrated in extensive simulation and multiple real data studies.
Source: arXiv (Cornell University)
Authors: Han, Xiao, Tong, Xin, Fan, Yingying
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3016550448