Algorithm of a knowledge extraction from neural network on the example of hot-dip zinc coating process

Algorithm of a knowledge extraction from neural network on the example of hot-dip zinc coating process

Sławomir Golak, Franciszek Grosman, Tadeusz Wieczorek

Politechnika Śląska.

DOI:

https://doi.org/10.7494/cmms.2003.2.0037

Abstract:

One of the major disadvantages of neural networks is that they can be considered as black boxes, since no satisfactory explanation of their work. A method for the analysis of regression neural networks, which provides physical interpretation of examined processes, is described in the paper. Method was verified on a base of test and empirical data of hot-dip zinc coating process.

Cite as:

Golak, S., Grosman, F., Wieczorek, T. (2003). Algorithm of a knowledge extraction from neural network on the example of hot-dip zinc coating process. Computer Methods in Materials Science, 3(2), 72 – 78. https://doi.org/10.7494/cmms.2003.2.0037

Article (PDF):

Keywords:

,

References:

Alan B. Tickle, R. Anders. (1998). The Truth Will Come to Light: Directions and Chanllengs in Extracting the Knowledge Embdded Withing Trained Artifical Neural Networks. IEEE Transactions on Neural Networks, vol. 9, no. 6, 1057-1068.

Arabas J. (2001). Wyktady z algorytmow ewolucyjnych. Wydawnictwo Naukowo-Techniczne, Warszawa. Grosman F, Wieczorek T., Golak S., 2001, Application of Neural Network for Analysis of Hot-Dip Zinc Coating Process Influence on Mechanical Properties of Metal Sheets, Sdchsische Fachtagung Umformtechnik, Freiberg, .

Grosman F., Wieczorek T., Golak S., 2002, Próba interpretacji procesu cynkowania ogniowego w oparciu o parametry struktury nauczonej sieci neuronowej, KomPlasTech X, Wisła, pp. 11-18.

Herrera F, Lozano M., Verdegay J. L., 1995, Generating Fuzzy Rules from Examples using Genetic Algorithms, Fuzzy Logic And Soft Computing, World Scientific, 93-98.

Marder A.R. (2000). A Review of the Metallurgy of Zine Coated Steel. Progress in Materials Science, vol. 45/3, pp. 191.

Osowski S. (1996). Sieci neuronowe w ujeciu algorytmicznym. Wydawnictwo Naukowo-Techniczne, Warszawa.

Saito K, Nakano R. (2002). Extracting regression rules from neural networks. Neural Networks, 15, 1279-1288.

Tanaka K., Nii M., Ishibuchi H. (1999). Learning from Linguistic Rules and Rule Etraction for Function Approximation by Neural Networks. SEAL’98, LNCS 1 585, .

Zhi-Hua Zhou, Yuan Jiang, Shi-Fu Chen, 2002, Extracting Symbolic Rules from Trained Neural Network Ensembles, A/ Communications, IOS Press.

Zurada J., Malinowski A., Usui S., 1997 Perturbation Method for Deleting Redundant Inputs of Perceptron Networks, Neurocomputing 14, pp. 177-193. Artykuł otrzymano 19 maja 2003 r., 317-324.