Abstract: Clustering is an important data mining application. One of the problems with traditional clustering methods is that they require user to define distance functions that are not always available. This paper presents a new method for clustering without distance functions. It is the combination of two basic methods. It has been tested on several well known data sets and the results revealed its validity.

摘要： 聚类分析是一种重要的数据挖掘运用。传统聚类方法的问题之一是它们需要用户定义距离函数，而所需的距离函数有时难以得到。文章介绍了一种新的无须距离函数的聚类方法。它是两种基本聚类方法的有机结合。对著名测试数据的试验表明这种方法是有效的。

关键词: 数据挖掘, 聚类, 并发最大化, 类标最小化

CLC Number: 

• TP311.12