摘要
该文针对RoughSet理论的属性约简进行了研究。利用RoughSet和信息论的相关知识,研究了通过可辨识矩阵求得属性约简集,并利用条件熵来计算属性约简集中属性间的相关性,其平均值最小的属性集即为求得的最佳属性约简的结果。实验证明,它可以取得比较理想的效果。最后利用该文的方法给出了对UCI机器学习数据库的例子的约简结果。
This paper discusses the attribute reduction in Rough Set theory.Using the knowledge of Rough Set and information theory,the paper gets attribute reduction set by discernibility matrix,and compute s relevance of attributes through conditional entropy.The best attribute reduction is the set which value is the minimum of average of relevance of attributes.The experiment shows that it can get better effectivity.Finally,it get reduction results of UCI databases using the algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2002年第5期15-19,78,共6页
Computer Engineering and Applications
基金
部级预研项目(编号:104050202(5))
