Inputs' Significance Analysis with Rough Sets Theory
The aim of this article is to show that a proper choice of a discretisation method is a key point in analysing the significance of attributes describing a specific real system of continuous attributes. In the second section of the article three most popular automatic discretisation methods are presented, which are next, in the third section, used for a discretisation of continuous attributes describing a system of an unemployment in Poland. The results obtained after application of these methods are compared with the results obtained with an expert method, proposed in the last part of the article.
KeywordsInputs' significance rough sets theory discretisation methods
Unable to display preview. Download preview PDF.
- Czerniak J., Zarzycki H., “Application of rough sets in the presumptive diagnosis of urinary system diseases”, Proceedings on 9th International Conference, ACS'2002, pp. 43–51, Kluwer Academic Publisher, London, 2003.Google Scholar
- Demuth H. Beale M., “Neural Network Toolbox User's Guide”, The Math Works Inc., Natick MA USA, 2000.Google Scholar
- Masters T., “Practical Neural Network Recipes in C++”, The Scientific-Technical Publishing House, Warsaw, 1996.Google Scholar
- Pelczar M., “Rough Sets Using for Medical Services Costs Modeling in Decision Support System”, 9th International Conference, ACS'2002, Informa, Szczecin, 2002.Google Scholar
- Rejer I., “The Method of Modeling the Multi-dimensional System with Artificial Intelligence Methods on the Example of an Unemployment in Poland”, The Scientific Publishing House of the Szczecin University, Szczecin, 2003.Google Scholar
- Rejer I., Piegat A., “A Method of Investigating a Significance of Input Variables in Non-Linear High-Dimensional Systems”, Proceedings on 9th International Conference, ACS'2002, pp. 73–80, Kluwer Academic Publisher, London, 2002.Google Scholar