作者搜索 |
![]() |
新闻 |
10.5.5C-DBLP系统正式发布作者研究兴趣及学术活动展示功能,请访问作者页面试用。
|
09.7.6C-DBLP的文献BibTex信息展示功能正式上线,请访问文章详细页面使用。
|
09.7.29C-DBLP系统新增同名区分功能,欢迎大家在作者页面试用。该功能部分使用了清华大学王建勇老师课题组提出的GHOST(GrapH-based framewOrk for name diStincTion)算法,在此表示感谢。
|
09.6.2C-DBLP系统集成了作者的相关图片并在搜索结果页面展示,敬请试用。
|
| 数值型和分类型混合数据的模糊K-Prototypes聚类算法(英文) BibTex | |
| 作者: | 陈宁 陈安 周龙骧 |
| 单位: | 中国科学院数学与系统科学研究院!北京100080;中国科学院科技政策与管理科学研究所!北京100080;中国科学院软件研究所;北京100080;中国科学院数学与系统科学研究院!北京100080 |
| 关键词: | 数值型属性;分类型属性;确定聚类;模糊聚类 |
| 出处: | 软件学报 2001 年 08期 |
| 基金: | 国家自然科学基金 No.6 99830 11&& |
| 全文链接: | 查看全文>> |
| 摘要: | |
| 由于数据库经常同时包含数值型和分类型的属性 ,因此研究能够处理混合型数据的聚类算法无疑是很重要的 .讨论了混合型数据的聚类问题 ,提出了一种模糊 K- prototypes算法 .该算法融合了 K- means和 K- modes对数值型和分类型数据的处理方法 ,能够处理混合类型的数据 .模糊技术体现聚类的边界特征 ,更适合处理含有噪声和缺失数据的数据库 .实验结果显示 ,模糊算法比相应的确定算法得到的结果准确度高 | |
| 正文快照: | |
| Clustering has been discussed extensively in many areas such as similarity search,customer segmentation,pattern recognition and trend analysis.The capacity to deal with both numeric and categorical valued attributes isundoubtedly important for clustering … | |
| Fuzzy K-Prototypes Algorithm for Clustering Mixed Numeric and Categorical Valued Data | |
| Author: | CHEN Ning 1; CHEN An 2;3; ZHOU Long xiang 1 1(Academy of Mathematics and System Sciences;The Chinese Academy of Sciences;Beijing 100080;China); 2(Institute of Policy and Management;The Chinese Academy of Sciences;Beijing 1 |
| Keywords: | numeric attribute;categorical attribute;hard clustering;fuzzy clustering |
| Abstract: | |
| The capacity of dealing with mixed numeric and categorical valued data is undoubtedly important for clustering algorithms because there is usually a mixture of numeric and categorical valued attributes in real databases. The use of fuzzy techniques makes clustering algorithms robust against noise and missing values in the databases. In this paper, a fuzzy kprototypes algorithm integrating k means and k modes algorithm is presented and is used to mixed databases. Experiments on several real databases demonstrate that fuzzy algorithm can get better result than the corresponding hard algorithm. Some properties of fuzzy k prototypes algorithm are also discussed. | |

10.5.5