A novel neural network technique for modelling data containing multiple functions
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.
KeywordsInput Space Digital Elevation Model Output Space Node Level Underlying Function
Unable to display preview. Download preview PDF.
- 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
- 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
- Zurada, JM: Introduction to Artificial Neural Systems, West Publishing Company (ISBN 0-314-93391-3) p58, 1992.Google Scholar
- 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