计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (23): 141-143.DOI: 10.3778/j.issn.1002-8331.2010.23.040

• 数据库、信号与信息处理 • 上一篇    下一篇

结合仿射几何和正交分解的类中心分类法研究

张玉芳1,梁 琪1,熊忠阳1,肖 婷2   

  1. 1.重庆大学 计算机学院,重庆 400030
    2.西南大学 计算机信息与科学学院,重庆 400030
  • 收稿日期:2009-01-15 修回日期:2009-03-27 出版日期:2010-08-11 发布日期:2010-08-11
  • 通讯作者: 张玉芳

Centroid-based classification based on affine geometry and orthogonal decomposition

ZHANG Yu-fang1,LIANG Qi1,XIONG Zhong-yang1,XIAO Ting2   

  1. 1.College of Computer Science,Chongqing University,Chongqing 400030,China
    2.School of Computer Information and Science,Xinan University,Chongqing 400030,China
  • Received:2009-01-15 Revised:2009-03-27 Online:2010-08-11 Published:2010-08-11
  • Contact: ZHANG Yu-fang

摘要: 类中心分类法是非常有效的分类算法。但由于向量空间产生空间扭曲,导致类中心分类法处理某些界限不明显的类别精确度不高。对此引入仿射几何和力的正交分解的方法,提出基于仿射几何和力的正交分解模型的类中心分类法的改进算法。在降低了计算维度的同时解决了由于向量空间特征项维被认为是正交带来的空间扭曲引起的误差。

关键词: 仿射几何, 正交分解, 类中心分类法

Abstract: Centroid-based classification is an effective text classification method.The spatial warping brought by vector space causes low precision when centroid-based classification method treats nonobvious boundary.This paper introduces affine geometry and orthogonal decomposition into centroid-based classification.This algorithm can reduce calculate dimensionality and error caused by spatial warping.

Key words: affine geometry, orthogonal decomposition, centroid-based classification method

中图分类号: