|Table of Contents|

[1] Wang Hongwei, Yi Lei, Wang Jianhui,. New text classification algorithmbased on interdependence and equivalent radius [J]. Journal of Southeast University (English Edition), 2007, 23 (1): 63-69. [doi:10.3969/j.issn.1003-7985.2007.01.014]
Copy

New text classification algorithmbased on interdependence and equivalent radius()
基于互依赖和等效半径的文本分类方法

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:
23
Issue:
2007 1
Page:
63-69
Research Field:
Automation
Publishing date:
2007-03-30

Info

Title:
New text classification algorithmbased on interdependence and equivalent radius
基于互依赖和等效半径的文本分类方法
Author(s):
Wang Hongwei1, Yi Lei2, Wang Jianhui3
1School of Economics and Management, Tongji University, Shanghai 200092, China
2Department of Mathematics, Fudan University, Shanghai 200433, China
3School of Information Science and Engineering, Fudan University, Shanghai 200433, China
王洪伟1, 伊磊2, 王建会3
1同济大学经济与管理学院, 上海 200092; 2复旦大学数学科学学院, 上海 200433; 3复旦大学信息科学与工程学院, 上海 200433
Keywords:
classification;  equivalent radius;  vector space;  interdependence;  interdependence and equivalent radius
分类;  等效半径;  向量空间;  互依赖;  IER
PACS:
TP139
DOI:
10.3969/j.issn.1003-7985.2007.01.014
Abstract:
To improve the traditional classifying methods, such as vector space model(VSM)-based methods with highly complicated computation and poor scalability, a new classifying method(called IER)is presented based on two new concepts:interdependence and equivalent radius.In IER, the attribute is selected according to the value of interdependence, and the classifying rule is based on equivalent radius and center of gravity.The algorithm analysis shows that IER is good at classifying a large number of samples with higher scalability and lower computation complexity.After several experiments in classifying Chinese texts, the conclusion is drawn that IER outperforms k-nearest neighbor(kNN)and classifcation based on the center of classes(CCC)methods, so IER can be used online to automatically classify a large number of samples while keeping higher precision and recall.
为了解决传统分类方法计算复杂度高及可扩展性差的问题, 提出了互依赖和等效半径的概念, 并将两者相结合, 提出新的分类算法——基于互依赖和等效半径、易更新的分类算法IER.IER算法根据互依赖作为特征选择的量度, 通过较长特征值的选择降低维度, 通过重心和等效半径来建立分类模型.算法分析显示IER计算复杂度较低, 扩展性能较好, 适用于大规模场合.将IER算法应用于中文文本分类, 并与kNN算法和类中心向量法进行比较, 结果表明, 在提高分类精度的同时, IER还可以大幅度提高分类速度, 有利于对大规模信息样本进行实时在线的自动分类.

References:

[1] Bian Zhaoqi, Zhang Xuegong.Pattern recognition [M].Beijing:Tsinghua University Press, 2001:9-43.(in Chinese)
[2] Burges C J C.A tutorial on support vector machines for pattern recognition [J].Data Mining and Knowledge Discovery, 1998, 2(2):955-974.
[3] Schapire R, Singer Y.BoosTexter:a boosting-based system for text categorization [J].Machine Learning, 2000, 39(2/3):135-168.
[4] Dasarathy Y.Minimal consistent set(MCS)identification for optimal nearest neighbor decision system terms design [J].IEEE Trans Syst Man Cybern, 1996, 24(3):511-517.
[5] Lam W, Ho C Y.Using a generalized instance set for automatic text categorization [C]//Proc of the 21th Ann Int ACM SIGIR Conference on Research and Development in Information Retrieval.Melbourne, Australia, 1998:81-89.
[6] Zhou Shuigeng.The research on Chinese text database and Chinese text processing [D].Shanghai:Fudan University, 2001.(in Chinese)
[7] Peng Fuchun, Schuurmans Dale.Self-supervised Chinese word segmentation [C]//Proc of the 4th Int Symposium on Intelligent Data Analysis.Cascais, Portugal, 2001:238-247.
[8] Sproat R, Shih C L A stochastic finite-state word segmentation algorithm for Chinese [J].Computational Linguistics, 1996, 22(3):377-404.
[9] Wang Jianhui, Hu Yunfa.An algorithm to classify documents based on equivalent radius. Technical report, No.021011346 [R].Shanghai:Fudan University, 2002.(in Chinese)
[10] Emerson Thomas.Segmenting Chinese in Unicode [C]//Proc of the 16th Int Unicode Conference.Amsterdam, Holland, 2000:1-10.

Memo

Memo:
Biography: Wang Hongwei(1973—), male, doctor, lecturer, hwwang@mail.tongji.edu.cn.
Last Update: 2007-03-20