基于Shadowed Sets的连续属性离散化

基于Shadowed Sets理论研究了粗糙集连续属性离散化问题,提出一种新的基于Shadowed Sets 理论的候选断点集提取算法.该算法根据实例在单属性上的分布,对数据样本进行分类,采用Shadowed Sets计算出各类的上下近似,最终提取出候选断点集.使用多组UCI数据对此算法的性能进行检验,同时还与其它候选断点集提取算法做了对比实验.实验结果表明,此算法能有效地减少数据集候选断点的数目,提高离散化算法运行速度和识别率.

Discretization of Continuous Attributes Based on Shadowed Sets

In this paper,discretization of continuous attributes in rough sets theory is researched and a new algorithm for extracting candidate cuts is proposed based on shadowed sets.According to distribution of the instance,the algorithm sorts the instance,and uses shadowed sets to calculate every class’s the upper and lower approximation,at last the candidate cuts are extracted.Several UCI data sets are applied to test the performance of the algorithm and the experiment results are compared with other algorithms.The experiment results show that the algorithm can effectively reduce the number of candidate cuts in data sets,and increase the speed of discretization and the rate of recognition.