基于网格的多密度聚类算法 |
| |
引用本文: | 张西芝,姬波,邱保志.基于网格的多密度聚类算法[J].微计算机信息,2005(26). |
| |
作者姓名: | 张西芝 姬波 邱保志 |
| |
摘 要: | 提出了一种多密度网格聚类算法GDD。该算法主要采用密度阈值递减的多阶段聚类技术提取不同密度的聚类,使用边界点处理技术提高聚类精度,同时对聚类结果进行了人工干预。GDD算法只要求对数据集进行一遍扫描。实验表明,该算法可扩展性好,能处理任意形状和大小的聚类,能够很好的识别出孤立点或噪声,在处理多密度聚类方面有很好的精度。
|
关 键 词: | 密度阈值递减 多阶段聚类 边界点提取 |
Grid-based Clustering Algorithm for Multi-density |
| |
Abstract: | This paper presents a grid- based clustering algorithm for multi- density(GDD). The GDD is a kind of the multi- stage clustering that integrates grid- based clustering, the technique of density threshold descending and border points extraction. Scan- ning the dataset only once, the GDD can discover clusters of ar- bitrary shapes. The experiment results show that it can discover outliers or noises effectively and get good cluster quality for multi- data sets. |
| |
Keywords: | density threshold descending multi- stage cluster- ing border points extraction |
|
|