A novel neural network technique for modelling data containing multiple functions

  • Owen M. Lewis
  • J. Andrew Ware
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1226)

Abstract

Increasingly neural network techniques are being applied to a wide range of pattern recognition and classification problems. However, there is often insufficient information available to facilitate optimal operation. This problem can lead to a situation where the data exhibits signs of containing multiple underlying functions. For example, if location is not included as a feature when modelling residential property appraisal, the data will appear to map across more than one underlying function. The methodology proposed in this paper uses a form of data stratification to overcome this problem. The premise followed is that it is better to produce multiple models that are specific to — and accurate within — certain scenarios, rather than a single model that is too general and therefore inaccurate.

Keywords

Input Space Digital Elevation Model Output Space Node Level Underlying Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Koncar N: Optimisation Methodologies for Direct Inverse Neurocontrol, Ph.D. Thesis, Department of Computing, 180 Queen's Gate London, SW7 2XZ, U.K., 1997Google Scholar
  2. Lewis OM, Ware JA, Jenkins DH: A Novel Neural Network Technique for the Valuation of Residential Property, Journal of Neural Computing and Applications, Springer Verlag, 1997.Google Scholar
  3. Zurada, JM: Introduction to Artificial Neural Systems, West Publishing Company (ISBN 0-314-93391-3) p58, 1992.Google Scholar
  4. Ware JA, Lewis OM, Kidner DB: A Neural Network Approach to the Compression of Digital Elevation Models, 5th GISRUK Research Conference — Leeds, 1997.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Owen M. Lewis
    • 1
  • J. Andrew Ware
    • 1
  1. 1.Division of Mathematics and ComputingUniversity of GlamorganPontypridd, Mid GlamorganUK

Personalised recommendations