一种面向混合属性数据聚类的新算法

New hybrid data orientation cluster algorithm

在分析传统聚类算法的基础上,提出一种针对混合属性数据的聚类算法.该算法利用格论中简单元组及超级元组将对象属性转化为格模型建立,以对象间格覆盖数来衡量类间相似度,根据高覆盖数高相似度的原则选择聚类中心进行聚类.在公共数据集上的实验结果表明,该算法在不增加空间复杂度的基础上,有效地提高了混合属性数据聚类的质量.

Based on the analysis of the traditional clustering algorithms, an algorithm is presented to cluster the hybrid data. The method changes the object＇s attributes to lattice based on the conception of simple tuples and hyper tuples, uses the numbers of covers to measure the similarity between labels, and chooses the clustering mean-point according to the rule of high covers to high similarity. Experiment results based on public data set show that the proposed algorithm can improve the quality of hybrid data clustering, and doesn＇t increase the space complexity.