An Ensemble of Neural Classifiers and Constructivist Algorithms in the Identification of Agricultural Suitability Complexes of Soils on the Basis of Physiographic Information

Table 2

Some properties of classifiers obtained from the Intelligent Problem Solver procedure. Acronyms: MLP: Multilayer Perceptron, RBF: Radial Basis Function, PNN: Probabilistic Neural Network, TPerf: identification correctness of the training part of the training T-1 set, VPerf: identification correctness of the validating part of the T-1c set, TePerf: identification correctness of the test part of the T-1 set.

Type of classifiers	Number of inputs	Number of hidden units	Range of values TPerf	Range of values VPerf	Range of values TePerf
MLP	1–16	2–24	0.34–0.75	0.34–0.73	0.34–0.73
RBF	8–16	10–72	0.45–0.61	0.45–0.61	0.46–0.60
PNN	16	10000	0.53–0.93	0.51–0.72	0.53–0.72