| 具有特征排序功能的鲁棒性模糊聚类方法 |
| |
| 引用本文： | 皋军,王士同.具有特征排序功能的鲁棒性模糊聚类方法[J].自动化学报,2009,35(2):145-153. |
| |
| 作者姓名： | 皋军 王士同 |
| |
| 作者单位： | 1.江南大学信息工程学院 无锡 214122 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划)，国家自然科学基金，浙江大学CAD＆CG国家重点实验室开放课题 |
| |
| 摘 要： | 提出了一种加权模糊聚类算法, 其优势在于能在实现有效聚类的同时, 对样本噪音进行识别和按样本特征对聚类的贡献程度进行排序. 因此, 本文所提出的方法具有鲁棒性, 并可对所得的特征排序进行特征选择, 实验结果表明了该方法具有上述优势.
|
| 关 键 词： | 模糊聚类 收敛性 权参数 鲁棒性 |
| 收稿时间： | 2007-6-18 |
| 修稿时间： | 2008-8-11 |
Fuzzy Clustering Algorithm with Ranking Features and Identifying Noise Simultaneously |
| |
| GAO Jun,WANG Shi-Tong.Fuzzy Clustering Algorithm with Ranking Features and Identifying Noise Simultaneously[J].Acta Automatica Sinica,2009,35(2):145-153. |
| |
| Authors: | GAO Jun WANG Shi-Tong |
| |
| Affiliation: | 1.School of Information Engineering, Jiangnan University, Wuxi 214122;2.School of Information Engineering, Yancheng Institute of Technology, Yancheng 224001;3.State Key Laboratory of CAD &CG, Zhejiang University, Hangzhou 310027 |
| |
| Abstract: | This paper proposes a weighted fuzzy clustering algorithm (FCA) that can identify the noise of samples and rank the samples' features simultaneously according to their contribution degrees, while realizing clustering efficiently. Therefore, the FCA is robust and can be used to extract the sample's features. Experimental results indicate the above advantages of the FCA. |
| |
| Keywords: | Fuzzy clustering convergence weights robustness |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
|
