| 摘要: |
| 详细分析了数据流聚类算法CluStream的不足之处,如对非球形的聚类效果不好、对周期性数据的聚类变化反映不完整等,并针对这些不足之处提出了一种采用空间分割、组合以及按密度聚类的算法ACluStream.实验结果表明,ACluStream在准确度和速度上都比CluStream有较大的提高. |
| 关键词: 数据流 聚类 数据挖掘 |
| DOI： |
| 分类号: |
| 基金项目:Supported by the National Natural Science Foundation of China under Grant No.60573097(国家自然科学基金);the Research Foundation of National Science and Technology Plan Project of China under Grant No.2004BA721A02(国家科技计划);the Research Foundation of Disciplines Leadingto Doctorate Degree of Chinese Universities under Grant No.20050558017(高等学校博士学科点专项科研基金);the Natural Science Foundation of Guangdong Province of China under Grant Nos.05200302,04300462(广东省自然科学基金);the Research Foundation of Science and Technology Plan Project in Guangdong Province of China under Grant No.2005B10101032(广东省科技计划项目) |
|
| Arbitrary Shape Cluster Algorithm for Clustering Data Stream |
|
ZHU Wei-Heng,YIN Jian,XIE Yi-Huang
|
| Abstract: |
| CluStream is a popular data stream cluster algorithm, however, it is not capable enough to cluster arbitrary shapes and make clusters in periodic data. This paper introduces a new algorithm ACluStream to solve these problems. The ACluStream is based on the partition and assemble of the space and cluster by density. In the experiment, it is shown that ACluStream is better than CluStream in speed and accuracy. |
| Key words: data stream clustering data mining |
