A0709
Title: Evaluating the error probability of the spectral clustering algorithm in the allometric extension model
Authors: Kohei Kawamoto - Kyushu University (Japan) [presenting]
Yuichi Goto - Kyushu University (Japan)
Koji Tsukuda - Kyushu University (Japan)
Abstract: The spectral clustering algorithm is often used as a binary clustering method for unclassified data by applying the principal component analysis. Several properties of the method have been studied under the assumption of the equality of two population covariance matrices. A non-asymptotic bound of the error probability of clustering is provided under the assumption of the allometric extension model; that is, the directions of the first eigenvectors of two covariance matrices and the direction of the difference of two mean vectors coincide.