通过粗糙集理论对一种实值属性约简算法进行了研究,给出了实值决策系统属性约简的算法,并采用UCI中的数据集进行分析,实验结果表明:该约简方法可以选择较少的属性而保持或改善分类能力.
The algorithm for numerical attribute reduction is studied by rough set theory.The algorithm for attribute reduction on decision information systems is presented.Furthermore,experimental results with UCI data sets show that this method can select a few attributes and keep or improve classification ability.