relative reduction

This algorithm adopted the relative Bhattacharyya distance summation of each feature as the criterion function to select the features which contributed more to the reduction of classification error. These features then could be used for Gaussian maximum likelihood classification.

该算法以各特征的相对Bhattacharyya和作为准则函数选择能有效降低分类错误率的一组特征，最后利用这组特征进行高斯似然分类。

Attribute reduction and relative attribute reduction are a core of KDD.

粗糙集理论中,属性约简是知识挖掘的核心.

With all condition attributes as the initial reduction, this algorithm takes relative positive-region as the iterative criterion to assure that classification ability of the resulted reduction does not decline.

该算法以相对正域为迭代准则，以所有条件属性为初始约简集合，通过逐步缩减来求取约简，保证了所求取的约简对问题的分类能力不会减弱。

Therefore, on the basis of attribute reduction of static database, the definition of relative difference comparison table has been given, and a new algorithm for minimum attribute reduction with conditional attribute has been brought forward.

因此，本文就是在静态数据库的属性约简的基础上，引入过渡相对差异比较表，并在此基础上提出了一种新的增加条件属性的最小属性约简的算法。

The equivalence among classification , relative positive-region and variable precision rough sets, attributes reduction becomes multiform.

并且三者之间打破了在经典粗糙集模型中的等价性,需要针对三者分别建立模型, 使属性约简变得多样化.