Program Realization of Rough Set Attributes Reduction
Xuefeng Zhang,Qingling Zhang
DOI: https://doi.org/10.1109/wcica.2006.1714230
2006-01-01
Abstract:Rough set theory is a new mathematical tool to deal with imprecise, incomplete and inconsistent data. First, the basic constitute of data analysis system based on rough set method is briefly described. Discretization method of continuous attributes is considered and continuous attributes are changed into discrete attributes. Two important concepts of indiscernibility relation and relatively positive regions are mainly focused on. By using the dependant degree of knowledge, the algorithm of rough set data analysis system is submitted. By being compared the numbers of reduced attributes, the result of minimal attributes reduction is gotten. Program realization of many algorithms of solving relative core, upper approximation, lower approximation, equivalence relation, relatively significant degree, relatively attributes reduction, relatively value reduction, minimal decision rules is obtained, respectively. The MATLAB programs of above fields are given. At last, running results of factual engineering system are promoted. Simulation results for pattern recognizing show that the method improves the rate of distinguish. It is factual significant in promoting application of rough set theory.