| 一种基于N-Gram改进的文本特征提取算法 |
| |
| 引用本文： | 于津凯,王映雪,陈怀楚.一种基于N-Gram改进的文本特征提取算法[J].图书情报工作,2004,48(8):48-50,43. |
| |
| 作者姓名： | 于津凯 王映雪 陈怀楚 |
| |
| 作者单位： | 清华大学计算机与信息管理中心, 北京 100084 |
| |
| 摘 要： | 介绍一种改进的文本特征提取及匹配算法。该算法基于N-Gram算法思路进行文本处理和特征提取，设计了gram关联矩阵用于统计与合并特征词，从而在固定长度N-Gram算法的基础上能够提取出不同长度的特征词。实验证明，该特征提取算法能够更为准确地描述文本特征，可应用于文本检索、Web挖掘等信息处理领域。
|
| 关 键 词： | 文本特征提取 N-Gram算法 gram关联矩阵 |
| 收稿时间： | 2003-11-03 |
An Improved Text Feature Extraction Algorithm Based on N-Gram |
| |
| Yu Jinkai Wang Yingxue Chen Huaichu
Computer & Information Management Center,Tsinghua University,Beijing.An Improved Text Feature Extraction Algorithm Based on N-Gram[J].Library and Information Service,2004,48(8):48-50,43. |
| |
| Authors: | Yu Jinkai Wang Yingxue Chen Huaichu
Computer & Information Management Center Tsinghua University Beijing |
| |
| Institution: | Computer & Information Management Center, Tsinghua University, Beijing 100084 |
| |
| Abstract: | This paper introduces an improved text feature extraction algorithm based on N - Gram theory. It designs a gram correlative matrix to unite the consecutive bigrams into a multigram and breaks the limit of N - Gram which has fixed - length gram extractions and forms the multigram features. |
| |
| Keywords: | text feature extraction N-Gram algorithm gram correlative matrix |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
