基于密度和混合距离度量方法的混合属性数据聚类研究

针对基于密度的传统算法不能处理混合属性数据,以及目前的混合属性聚类算法大多数聚类质量不高等问题,提出了基于密度和混合距离度量方法的混合属性聚类算法.该算法通过分析混合属性数据特征,将混合属性数据分为数值占优、分类占优和均衡型混合属性数据3类,分析不同情况的特征选取相应的距离度量方式,通过预设参数能够发现数据密集区域,确定核心点,再利用核心点确定密度相连的对象实现聚类,获得最终的聚类结果.将算法应用于多种数据集上的实验结果表明,该算法具有较高的聚类质量,能够有效处理混合属性数据.

Density-based clustering algorithm for numerical and categorical data with mixed distance measure methods

Traditional density-based clustering algorithm cannot deal with the mixed data, and the accuracy of most existing clustering algorithms for mixed data is not high enough as desired. To solve the problem, a density-based clustering algorithm for mixed data with mixed distance measure is proposed. Firstly, the characteristics of the mixed attribute data are analyzed, and then the data is divided into three parts: numerical dominant data, categorical dominant data and balanced data. According to the situation of dominance, corresponding distance measure method is selected. Distance between objects is calculated for finding the dense regions, and core objects are defined by preset parameters. Then, by making use of the core points to determine the objects with neighboring densities to form clusters, we obtain the final clustering result. Experiments on real data sets show that the algorithm can achieve better clustering results, and can deal with the numerical and categorical data efficiently.