Abstract: In the research and development of intelligence system，clustering analysis is a very important problem.A formula for similarity measures based on unknown degree and core between Vague sets is presented.Under the consideration of algorithm independence and computing complexity，benefiting from the idea of clustering analysis at fuzzy sets，a new direct clustering algorithm using similarity measure of Vague sets as evaluation criteria is presented.Using the formula，the Vague transfer closure method and the Vague direct clustering method are used to calculate respectively.The experimental result shows that the direct clustering method based on the similarity of Vague sets is easy，not causing distortion of the original information，and there is no special requirement about the size of the amount of data at the same time.It is more effective than Vague transfer closure method.

Key words: Vague sets, similarity measure, clustering analysis, Fuzzy sets, direct clustering method

摘要： 在智能系统的研究与开发中，聚类分析是一个非常重要的问题。提出了一个基于未知度和核的Vague集间的相似度量公式。在考虑算法自主性和计算复杂性的基础之上，通过参考Fuzzy集中的相关聚类分析方法，给出了一种以Vague集的相似度量为评价准则的直接聚类算法。使用相似度量公式，分别采用Vague 传递闭包法和Vague 直接聚类法进行计算，实验结果表明，基于Vague 相似度量的直接聚类法计算简单，不会造成原始信息的失真，而且对数据量的大小均无特别的要求，比Vague 传递闭包法更加有效。

关键词: Vague集, 相似度量, 聚类分析, Fuzzy集, 直接聚类法