基于DF关系,给出了DF关系截矩阵的定义,以及由DF相似关系构造DF等价关系的方法,并在此基础上结合数据挖掘中的分类技术,提出了一种面向DF数据的聚类算法,该算法的提出不但能拓宽聚类对象的范围,而且更符合实际需求。最后给出了运用该算法的示例。
Based on dynamic fuzzy relation,the concept of cut-matrix and a method of building dynamic fuzzy equivalence relation according to dynamic fuzzy similarity relation are presented.On such basis,combining classification technology in data mining,a clustering algorithm for dynamic fuzzy data is proposed.Using this algorithm,not only widen the range of clustering objects,but also can meet actual requirement.At last,an example is presented to illustrate the proposed algorithm.