| 新颖的无监督特征选择方法 |
| |
| 引用本文： | 朱颢东,李红婵,钟勇. 新颖的无监督特征选择方法[J]. 电子科技大学学报(自然科学版), 2010, 39(3): 412-415. DOI: 10.3969/j.issn.1001-0548.2010.03.019 |
| |
| 作者姓名： | 朱颢东 李红婵 钟勇 |
| |
| 作者单位： | 1.郑州轻工业学院计算机与通信工程学院 郑州 450002; |
| |
| 基金项目： | 四川省科技计划项目，四川省科技攻关项目 |
| |
| 摘 要： | 针对有监督特征选择方法因为需要类信息而无法应用于文本聚类的问题，提出了一种新的无监督特征选择方法:结合文档频和K-Means的特征选择方法。该方法首先使用文档频进行无监督特征初选，然后再通过在不同K-Means聚类结果上使用有监督特征选择方法来实现无监督特征选择。实验表明该方法不仅能够成功地选择出最为重要的—小部分特征，而且还能提高聚类质量。
|
| 关 键 词： | 分类 聚类算法 文档频 特征选择 K-Means |
| 收稿时间： | 2008-09-17 |
New Unsupervised Feature Selection Method |
| |
| ZHU Hao-dong,LI Hong-chan,ZHONG Yong. New Unsupervised Feature Selection Method[J]. Journal of University of Electronic Science and Technology of China, 2010, 39(3): 412-415. DOI: 10.3969/j.issn.1001-0548.2010.03.019 |
| |
| Authors: | ZHU Hao-dong LI Hong-chan ZHONG Yong |
| |
| Affiliation: | 1.School of Computer and Communication Engineering,Zhengzhou University of Light Industry Zhengzhou 450002;2.Chengdu Institute of Computer Application,Chinese Academy of Sciences Chengdu 610041;3.The Graduate School of the Chinese Academy of Sciences Shijingshan Beijing 100039 |
| |
| Abstract: | Due to unavailability of class label information, supervised feature selection methods can not be applied to text clustering. In this case, a new unsupervised feature selection method combined Document Frequency with K-Means is proposed. The method firstly employs document frequency to select initial unsupervised features, and then brings into unsupervised feature selection by means of mainly performing effective supervised feature selection methods on different K-Means clustering results. Experimental results show that the new method can not only successfully select out the best small part of features, but also can significantly improve clustering performance. |
| |
| Keywords: | K-Means |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《电子科技大学学报(自然科学版)》浏览原始摘要信息 |
|
点击此处可从《电子科技大学学报(自然科学版)》下载全文 |
