|
Descripción
|
|
|---|---|
| This paper proposes a method in order to detect the importance of the input variables in multivariate analysis problems. When there is correlation among predictor variables, the importance of each input variable, when adding variables in the model, can be detected from the knowledge stored in Artificial Neural Network (NN) and it must be taken into account. Neural networks models have been used with the analysis of sensibility, these models predict more accurately the relationship between variables, and it is the way to find a set of forecasting variables in order to be included in the new prediction model. The obtained results have been applied in a system to forecast the volume of wood for a tree, and to detect relationships between input and output variables | |
|
Internacional
|
Si |
|
JCR del ISI
|
No |
|
Título de la revista
|
Journal of Mathematics and System Science |
|
ISSN
|
2159-5291 |
|
Factor de impacto JCR
|
|
|
Información de impacto
|
|
|
Volumen
|
1 |
|
DOI
|
|
|
Número de revista
|
|
|
Desde la página
|
12 |
|
Hasta la página
|
22 |
|
Mes
|
MAYO |
|
Ranking
|
|