基于模糊粗糙集的一种属性约简算法

将粗糙集理论与模糊集理论相结合,提出了一种基于模糊粗糙集的属性约简算法。该方法引入了模糊C均值聚类算法用以连续属性的模糊化;并通过聚类有效性分析来确定最佳分类数目;克服了目前属性模糊化方法需要人为规定划分类数,几乎不考虑信息系统的具体属性值等缺点。实例验证了此方法的有效性。

An Attribute Reduction Approach Based on Fuzzy-Rough Set

A new method of attribute reduction based on fuzzy-rough set is proposed. In view of the deficiency of attribute fuzzified method, the fuzzy C means clustering is introduced in the article in order to fuzzify the continual attribute, and the best minute class number is obtained by the valid analysis of clustering. At last, examples are illustrated and proved that the approach is effective.