EFFICIENCY OF MULTILAYERED FEEDFORWARD NEURAL NETWORKS ON CLASSIFICATION IN RELATION TO LINEAR DISCRIMINANT-ANALYSIS, QUADRATIC DISCRIMINANT-ANALYSIS AND REGULARIZED DISCRIMINANT-ANALYSIS
Ms. Sanchez et La. Sarabia, EFFICIENCY OF MULTILAYERED FEEDFORWARD NEURAL NETWORKS ON CLASSIFICATION IN RELATION TO LINEAR DISCRIMINANT-ANALYSIS, QUADRATIC DISCRIMINANT-ANALYSIS AND REGULARIZED DISCRIMINANT-ANALYSIS, Chemometrics and intelligent laboratory systems, 28(2), 1995, pp. 287-303
The efficiency of multi-layered feed-forward networks (MLF) on classif
ication is evaluated by applying them to simulated data. The classes a
re normal multivariate with three different structures for the matrix
of covariance. For each of them a complete factorial design, 2(3), was
performed, with a replicated central point in order to study the effe
ct of the relationships objects-variables, noise-signal and distance b
etween centroids. The results were compared to those obtained by apply
ing linear discriminant analysis, quadratic discriminant analysis and
regularized discriminant analysis to the same sets of data. The compar
ison was carried out by an ANOVA of the experimental designs and by pr
incipal components and correspondence analysis.