| 一种改进的孤立点消除及网络文本聚类算法 |
| |
| 引用本文： | 罗姗姗,杜庆治,杨秋萍,龙华.一种改进的孤立点消除及网络文本聚类算法[J].云南师范大学学报(自然科学版),2012,32(1):48-51. |
| |
| 作者姓名： | 罗姗姗 杜庆治 杨秋萍 龙华 |
| |
| 作者单位： | 昆明理工大学信息工程与自动化学院,云南昆明,650051 |
| |
| 基金项目： | 云南省科技厅社会发展专项资助项目(2007CA027) |
| |
| 摘 要： | 通过对CURE（Clustering Using Representatives）的研究,在此基础上,针对网络话题文本内容广泛、孤立点较多的特点,增加对孤立点的预处理过程,提出一种适用于网络话题文本聚类的层次聚类算法。该算法能降低算法对孤立点的敏感度,同时提高聚类效率,通过实验证明改进后的算法更适合运用在网络话题文本的快速聚类中。
|
| 关 键 词： | 层次聚类 CURE 孤立点 |
Improved Handling Outliers Method and Clustering Algorithm for Network Text |
| |
| LUO Shan-shan
,
DU Qing-zhi
,
YANG Qiu-ping
,
LONG Hua.Improved Handling Outliers Method and Clustering Algorithm for Network Text[J].Journal of Yunnan Normal University (Natural Sciences Edition),2012,32(1):48-51. |
| |
| Authors: | LUO Shan-shan DU Qing-zhi YANG Qiu-ping LONG Hua |
| |
| Institution: | (Information Engineering and Automation College,Kunming University of Science and Technology,Kunming 650500,China) |
| |
| Abstract: | Based on the research of CURE,a hierarchical Clustering Algorithm for Network texts which have a wide range and include a great deal of outliers has been proposed.This algorithm which have been added a pretreatment process for outliers can reduce the sensitivity of outliers,while improving the clustering efficiency.The experiments show that the improved algorithm is more suitable for fast clustering on network texts. |
| |
| Keywords: | Hierarchical Clustering CURE Outliers |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
