Supervised Discretization for Rough Sets – a Neighborhood Graph Approach

Archive - Central European Conference on Information and Intelligent Systems, CECIIS - 2011

Visnja Ognjenovic, Vladimir Brtka, Martin Jovanovic, Ivana Berkovic

Last modified: 2011-10-03

Abstract

Rough set theory has become an important mathematical tool for dealing with uncertainty in data. The data discretization is one of the main problems to be solved in the process of synthesis of decision rules from table-organized data. In this paper, we present a new discretization method in the context of supervised training. This method is based on the neighborhood graph. To evaluate supervised discretization, we used data sets obtained from the UCI Machine Learning Repository. We have used the Rosetta system and proposed SSCO system. The experimental results show that our method is effective.

Full Text: PDF