White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

A binary neural k-nearest neighbour technique

Hodge, V.J. and Austin, J. (orcid.org/0000-0001-5762-8614) (2005) A binary neural k-nearest neighbour technique. Knowledge and Information Systems. pp. 276-292. ISSN 0219-3116

Text (hodgevj3.pdf)

Download (686Kb)


K-Nearest Neighbour (k-NN) is a widely used technique for classifying and clustering data. K-NN is effective but is often criticised for its polynomial run-time growth as k-NN calculates the distance to every other record in the data set for each record in turn. This paper evaluates a novel k-NN classifier with linear growth and faster run-time built from binary neural networks. The binary neural approach uses robust encoding to map standard ordinal, categorical and real-valued data sets onto a binary neural network. The binary neural network uses high speed pattern matching to recall the k-best matches. We compare various configurations of the binary approach to a conventional approach for memory overheads, training speed, retrieval speed and retrieval accuracy. We demonstrate the superior performance with respect to speed and memory requirements of the binary approach compared to the standard approach and we pinpoint the optimal configurations.

Item Type: Article
Copyright, Publisher and Additional Information: Copyright © 2005 Springer-Verlag. This is an author produced version of a paper published in Knowledge and Information Systems. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.The original publication is available at www.springerlink.com.
Keywords: binary neural network,correlation matrix memory,K-nearest neighbour,parabolic kernel
Institution: The University of York
Academic Units: The University of York > Computer Science (York)
Depositing User: Sherpa Assistant
Date Deposited: 04 Nov 2005
Last Modified: 23 May 2016 00:23
Published Version: http://dx.doi.org/10.1007/s10115-004-0191-4
Status: Published
Refereed: Yes
URI: http://eprints.whiterose.ac.uk/id/eprint/766

Actions (repository staff only: login required)