Document classification method with small training data

Yasunari Maeda; Hideki Yoshida; Toshiyasu Matsushima

Document classification method with small training data

Yasunari Maeda^*, Hideki Yoshida, Toshiyasu Matsushima

^*Corresponding author for this work

School of Fundamental Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Document classification is one of important topics in the field of NLP(Natural Language Processing). In our previous research we've proposed a document classification method which minimizes an error rate with reference to a Bayes criterion. But when the number of documents in training data is small, the accuracy of the previous method is low. So in this research we propose a document classification method whose accuracy is higher than the previous method when the number of documents in training data is small.

Original language	English
Title of host publication	ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings
Pages	138-141
Number of pages	4
Publication status	Published - 2009 Dec 1
Event	ICROS-SICE International Joint Conference 2009, ICCAS-SICE 2009 - Fukuoka, Japan Duration: 2009 Aug 18 → 2009 Aug 21

Publication series

Name	ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings

Other

Other	ICROS-SICE International Joint Conference 2009, ICCAS-SICE 2009
Country/Territory	Japan
City	Fukuoka
Period	09/8/18 → 09/8/21

Keywords

Document classification
Estimating data
Prior distributions
Small training data

ASJC Scopus subject areas

Information Systems
Control and Systems Engineering
Industrial and Manufacturing Engineering

Cite this

Document classification method with small training data. / Maeda, Yasunari; Yoshida, Hideki; Matsushima, Toshiyasu.

ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings. 2009. p. 138-141 5333327 (ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Maeda, Y, Yoshida, H & Matsushima, T 2009, Document classification method with small training data. in ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings., 5333327, ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings, pp. 138-141, ICROS-SICE International Joint Conference 2009, ICCAS-SICE 2009, Fukuoka, Japan, 09/8/18.

@inproceedings{b2c142d21f214b63a2557bb48254cc8e,

title = "Document classification method with small training data",

abstract = "Document classification is one of important topics in the field of NLP(Natural Language Processing). In our previous research we've proposed a document classification method which minimizes an error rate with reference to a Bayes criterion. But when the number of documents in training data is small, the accuracy of the previous method is low. So in this research we propose a document classification method whose accuracy is higher than the previous method when the number of documents in training data is small.",

keywords = "Document classification, Estimating data, Prior distributions, Small training data",

author = "Yasunari Maeda and Hideki Yoshida and Toshiyasu Matsushima",

year = "2009",

month = dec,

day = "1",

language = "English",

isbn = "9784907764333",

series = "ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings",

pages = "138--141",

booktitle = "ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings",

note = "ICROS-SICE International Joint Conference 2009, ICCAS-SICE 2009 ; Conference date: 18-08-2009 Through 21-08-2009",

}

TY - GEN

T1 - Document classification method with small training data

AU - Maeda, Yasunari

AU - Yoshida, Hideki

AU - Matsushima, Toshiyasu

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Document classification is one of important topics in the field of NLP(Natural Language Processing). In our previous research we've proposed a document classification method which minimizes an error rate with reference to a Bayes criterion. But when the number of documents in training data is small, the accuracy of the previous method is low. So in this research we propose a document classification method whose accuracy is higher than the previous method when the number of documents in training data is small.

AB - Document classification is one of important topics in the field of NLP(Natural Language Processing). In our previous research we've proposed a document classification method which minimizes an error rate with reference to a Bayes criterion. But when the number of documents in training data is small, the accuracy of the previous method is low. So in this research we propose a document classification method whose accuracy is higher than the previous method when the number of documents in training data is small.

KW - Document classification

KW - Estimating data

KW - Prior distributions

KW - Small training data

UR - http://www.scopus.com/inward/record.url?scp=77951142189&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951142189&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:77951142189

SN - 9784907764333

T3 - ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings

SP - 138

EP - 141

BT - ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings

T2 - ICROS-SICE International Joint Conference 2009, ICCAS-SICE 2009

Y2 - 18 August 2009 through 21 August 2009

ER -

Document classification method with small training data

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Other files and links

Cite this