k-nearest neighbour (k-nn) model is a simple, popular classifier. Probabilistick-nn is a more powerful variant in which the model is cast in a Bayesian framework using (reversible jump) Markov chain Monte Carlo methods to average out the uncertainy over the model parameters.

Thek-nn classifier depends crucially on the metric used to determine distances between data points. However, scalings between features, and indeed whether some subset of features is redundant, are seldom knowna priori. Here we introduce a variable metric extension to the probabilistick-nn classifier, which permits averaging over all rotations and scalings of the data. In addition, the method permits automatic rejection of irrelevant features. Examples are provided on synthetic data, illustrating how the method can deform feature space and select salient features, and also on real-world data.

Unable to display preview. Download preview PDF.

Authors and Affiliations

1. Department of Computer Science, University of Exeter, UK

   Richard M. Everson & Jonathan E. Fieldsend

Editors and Affiliations

1. School of Engineering, Computing, and Mathematics, University of Exeter, EX4 4QF, Exeter, UK

   Zheng Rong Yang

2. School of Electrical and Electronic Engineering, University of Manchester, UK

   Hujun Yin

3. School of Engineering, Computer Science and Mathematics, University of Exeter, EX4 4QF, UK

   Richard M. Everson

Reprints and permissions

© 2004 Springer-Verlag Berlin Heidelberg

Policies and ethics