Interactive Classification through Neural Networks
In this paper, we present a simple and fast way to provide the operator a plane representation of multidimensional data through neural networks for interactive classification. The superiority of humans over automatic clustering procedures comes from their ability in recognising cluster structures in a two dimensional space, even in the presence of outliers between the clusters, of bridging clusters and of all kinds of irrelevant details in the data points distribution. When giving the operator the interactive means which will help him to isolate clusters of two dimensional points, this visualisation becomes base of a clustering procedure where the operator doesn’t loose his grip on the data he is analysing.
KeywordsHide Layer Multidimensional Data Plane Representation Dimensional Point Interactive Classification
Unable to display preview. Download preview PDF.
- Chien, Y., (1978) Interactive pattern recognition. New York: Marcel Dekker.Google Scholar
- Le Cun, Y. (1986) Learning process in asymmetric threshold network. Disorder systems and biological Organisation, E. Bienenstock, F. Fogelman Soulie, and G. Weibush (Eds), Berlin: Springer.Google Scholar
- Lippman, R.P. (1987) An introduction to computing with neural nets. IEEE ASSP Magazine, pp 4-22Google Scholar
- Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986) Learning internal representation by error propagation Parallel Distributed Processing: Explorations in the micro structures of cognition, Vol. 1, pp. 318–362, MIT Press, Cambridge, Mass.Google Scholar
- Wasserman, P.D. (1989) Neural computing Theory and Practise. VNR, New York.Google Scholar
- Cottrel, G.W., Munro, P.W. and Zipser, D. (1988) Image compression by back propagation: a demonstration of extensional programming. In: Advances in cognitive Science, Vol. 2, Sharkey, N. E. ed. Norwood, NJ Abbex.Google Scholar
- Baldi, P., Hornik, K. (1991) Backpropagation and unsupervised Learning in Linear Networks. Technical report, Jet Propulsion Laboratory and Division of Biology, California Institute of Technology.Google Scholar