Title :
Feedforward Bayesian neural network and continuous attributes
Author_Institution :
Fac. of Electr. & Comput. Eng., Ljubljana Univ.
Abstract :
Two methods for dealing with continuous attributes are proposed. The fuzzy learning method assumes fuzzy bounds of a continuous attribute during learning and the fuzzy classification method assumes fuzzy bounds during classification. The performance was tested on two medical diagnostic problems. The results obtained show that the proposed methods for dealing with continuous attributes perform better than the splitting of the attribute´s values with exact bounds
Keywords :
fuzzy logic; fuzzy set theory; learning systems; neural nets; pattern recognition; continuous attributes; feedforward Bayesian neural network; fuzzy bounds; fuzzy classification; fuzzy learning method; fuzzy logic; fuzzy set theory; medical diagnostic problems; pattern recognition; Backpropagation algorithms; Bayesian methods; Classification tree analysis; Computer networks; Feedforward neural networks; Learning systems; Machine learning algorithms; Medical diagnosis; Neural networks; Robustness;
Conference_Titel :
Neural Networks, 1991. 1991 IEEE International Joint Conference on
Print_ISBN :
0-7803-0227-3
DOI :
10.1109/IJCNN.1991.170395
