Data mining is currently a hot research topic of the knowledge discovery Domain.This paper mainly studies the way from data reduction Based on Rough Sets to data mining.Value reduction which generalizes every rule makes all rules more concise.The most important is it deduces the number of data, Then it improves the efficiency of data mining and deduces the complex of time .This paper gives method of data reduction Based on Roughsets.Implementation of concept Hierarchy.and a method for computing reduction using indiscernible matrix..Based on these theoretical work,we design and implement a prototype KDD system which is divided into four sections:preprocessing section,data reduction section,rule induction and decision algorithm section. This prototype system implements above methods and can test validity and efficiency of algorithm.
|