基于模糊度的计算几何分类器权重分配

为了充分发挥计算几何组合分类器的可视化特性,从类空间类别分布模糊特性出发,提出一种基于类空间模糊度的权重分配方法.该方法首先将子分类器由空间的类别表示转变为类别的空间表示,进而利用粗糙集理论分析各空间的模糊度,并将其作为该子分类器的权重.实验表明,利用模糊度信息进行加权后的分类器不但能与可视化特性更好地吻合,增强分类过程的可理解性,而且在分类精度上得到了进一步的提升,扩展了应用领域.

Weight calculation for computational geometry combining classifier using fuzzy of class space

In order to take advantage of visualization characteristics of computational geometry combined classifiers, a
weight calculation method based on space fuzzy is proposed according to the category distribution in class space. In this
method, the space is turned from the category information in space to the space information in category. The space fuzzy
is obtained from the later based rough set theory and taken as the weight for the sub classifier. The experiments show that
the classifier weighted by using the fuzzy degree can enhance the visual performance and the classify performance of the
classifier. It means that the comprehensibility of the classifier is enhanced and the application of the classifier is extended.