Practical neural network recipes in C++
Practical neural network recipes in C++
Feature Selection for Knowledge Discovery and Data Mining
Feature Selection for Knowledge Discovery and Data Mining
| Hi-index | 0.00 |
This article introduces a new method of investigating a significance of input variables in non-linear multi-dimensional systems. The main advantage of the proposed method, in comparison to other widely used methods, is that it allows to investigate the significance of input variables in larger subspaces of analysed system. The article presents not only theoretical basis of the proposed method but also a practical example which aim is to build a ranking of significance for a 19-dimensional system of an unemployment rate in Poland in years 1992 - 1999.