ORIGINAL PAPER
The artificial neural nerwork as a helping tool in the process of non-linear data compression
 
More details
Hide details
 
Journal of Research and Applications in Agricultural Engineering 2006;51(1):37-40
 
KEYWORDS
ABSTRACT
An autoassociative network is one which reproduces its inputs as outputs. Autoassociative networks have at least one hidden layer with less units than the input and output layers (which obviously have the same number of layers as each other). Hence, autoassociative networks perform some sort of dimensionality reduction or compression on the cases. Dimensionality reduction can be used to pre-process the input data to encode Information in a smaller number of variables. This approach recognizes that the intrinsic dimensionality of the data may be lower than the number of variables. In other words, the data can be adequately described by a smaller number of variables, if the right transformation can be found.
REFERENCES (8)
1.
Bishop C., (1995). Neural Networks for Pattern Recognition. Oxford: University Press.
 
2.
Fausett L., (1994). Fundamentals of Neural Networks. New York: Prentice Hall.
 
3.
Bouland H., Kamp Y., (1988). Auto-association by multilayer perceptrons and singular value decomposition. Biological Cybernetics 59, 291-294.
 
4.
Kramer M.A, (1991). Nonlinear principal components analysis using autoassociative neural networks. AlChe Journal 37 (2), 233-243.
 
5.
Osowski S. (2000). Sieci neuronowe do przetwarzania informacji: Oficyna Wydawnicza Politechniki Warszawskiej, Warszawa.
 
6.
Stateczny A., Praczyk T. (2002). Sztuczne sieci neuronowe w rozpoznawaniu obiektów morskich: GTN, Gdynia.
 
7.
Tadeusiewicz R. (1993). Sieci neuronowe: Akademicka Oficyna Wydawnicza, Warszawa.
 
8.
Boniecki P., Piekarska-Boniecka H.(2004). Neuronowa identyfikacja wybranych szkodników drzew owocowych w oparciu o analizę obrazu: Journal of Research and Applications in Agricultural Engineering, Vol. 49(3), str.25-30.
 
eISSN:2719-423X
ISSN:1642-686X
Journals System - logo
Scroll to top