一种基于二型模糊集的自适应熵模糊聚类

摘要

传统的模糊聚类算法在处理具有高阶模糊类别属性的数据集时无法准确对数据集进行建模.文章结合二型模糊集和基于熵的模糊聚类这两种概念,提出了一种二型模糊聚类算法.该算法通过一种新的自适应降型算法对二型模糊集进行降型,降低了二型模糊聚类算法的聚类的运算量.对比模糊聚类和基于熵的模糊聚类,该聚类算法在类别边界不够清晰的情况下,可以得到更好的聚类效果.%Traditional fuzzy clustering algorithms fail to deal with data sets with high-level fuzzy uncertainty.To solve this problem,this paper put together the type-2 fuzzy sets and entropy fuzzy clustering to propose a new type-2 fuzzy clustering algorithm.A noveltype reduction method via an adaptive parameter was proposed in this paper to improve the performance of type-2 clustering.The experimental results showed that this new type-2 fuzzy clustering algorithm outperformed the traditional algorithms when a great uncertainty existed in the boundaries of data sets.