PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Gaussian process latent variable models for fault detection
Luka Eciolaza, M. Alkarouri, Neil Lawrence, V. Kadirkamanathan and Peter Fleming
In: CIDM 2007, 1-5 April 2007, Honolulu, Hawaii, USA.

Abstract

The Gaussian process latent variable model (GPLVM) is a novel unsupervised approach to nonlinear low dimensional embedding proposed by Lawrence (2005). This paper presents the development of a framework for the implementation of the GPLVM for fault detection. A series of experiments have been carried out comparing and combining the GPLVM to the conventional and widely used linear dimension reduction technique of principal component analysis (PCA). The inclusion of the GPLVM for the visualisation and data analysis, led to a considerable improvement in the classification results

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:3810
Deposited By:Neil Lawrence
Deposited On:25 February 2008