Establishing the method of rough set clustering

传统的粗糙集理论关注的是决策表的监督分类，除了少数非监督分类的建议，也就是与粗糙集相关的聚类。然而，在本研究之前，有许多聚类技术没有被应用于粗糙集。我们研究了粗糙集聚类的理论背景，提出了包括层次聚类和非层次聚类在内的一系列新算法。此外，我们还将开发的方法应用于一些数据集，包括医疗信息中的数据集。因此，我们对粗糙集的聚类技术及其相关应用进行了广泛的概述。

Traditional rough set theory is concerned with supervised classification of decision tables, with the exception of a few proposals of unsupervised classification, in other words, clustering related to rough sets. However, there are many clustering techniques that had not been applied to rough sets before this study. We have studied theoretical background of clustering for rough sets, and developed a number of new algorithms including both hierarchical and non-hierarchical clustering. We moreover applied the developed methods to a number of data sets including those in medical information. We thus have shown a broad overview of clustering techniques of rough sets and related applications.