计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (22): 1-3.DOI: 10.3778/j.issn.1002-8331.2009.22.001

• 博士论坛 • 上一篇    下一篇

面向文档分类的LDE和简化SVM方法研究

王自强1,钱 旭1,孔 敏2

  

  1. 1.中国矿业大学(北京) 机电与信息工程学院,北京 100083
    2.山东省曲阜市职业中等专业学校,山东 曲阜 273100
  • 收稿日期:2009-04-15 修回日期:2009-05-18 出版日期:2009-08-01 发布日期:2009-08-01
  • 通讯作者: 王自强

LDE and simplified SVM method for document classification

WANG Zi-qiang1,QIAN Xu1,KONG Min2

  

  1. 1.College of Mechanical Electronic and Info. Eng.,China University of Mining and Technology(Beijing),Beijing 100083,China
    2.Qufu Vocational School of Shandong Province,Qufu,Shandong 273100,China
  • Received:2009-04-15 Revised:2009-05-18 Online:2009-08-01 Published:2009-08-01
  • Contact: WANG Zi-qiang

摘要: 为了快速准确地对文档进行分类,提出了一种基于局部鉴别嵌入LDE和简化SVM的高效文档分类算法。该算法首先利用LDE算法把高维文档数据投影到低维特征空间,然后在低维特征空间利用精简SVM进行分类。实验结果表明该算法具有分类准确率高和运行速度快的优点。

关键词: 文档分类, 局部鉴别嵌入, 简化支持向量机, 数据挖掘

Abstract: To rapidly and accurately perform document classification task,an efficient document classification algorithm is proposed by using Local Discriminant Embedding(LDE) and Simplified Support Vector Machine(SSVM) in this paper.The essential idea of the algorithm is as follows.The high dimensional document data are first projected into the lower dimensional feature space with LDE algorithm,then the SSVM algorithm is applied to classify documents in the lower dimensional feature space.Experimental results show that the proposed algorithm is of higher classification accuracy and faster running speed.

Key words: document classification, local discriminant embedding, simplified support vector machine, data mining