DocumentCode
1808052
Title
Using factorial design to optimise neural networks
Author
Santos, Marizete Silva ; Ludermir, Teresa Bernarda
Author_Institution
Univ. Fed. de Pernambuco, Recife, Brazil
Volume
2
fYear
1999
fDate
36342
Firstpage
857
Abstract
This work presents a statistical method (factorial design) to help in the design and implementation of a neural network. With the experiments that we have performed using this method it was possible to observe the influence of some parameters on network perfomance. Consequently, we were able to minimize the total number of simulations necessary to obtain a satisfactory result. For the example presented we have used a database of aroma signals from an artificial nose
Keywords
neural nets; optimisation; statistical analysis; aroma signal database; artificial nose; factorial design; neural network optimisation; simulation minimization; statistical method; Artificial neural networks; Data analysis; Design methodology; Design optimization; Humans; Neural networks; Neurons; Nose; Performance analysis; Statistical analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location
Washington, DC
ISSN
1098-7576
Print_ISBN
0-7803-5529-6
Type
conf
DOI
10.1109/IJCNN.1999.831064
Filename
831064
Link To Document