System evaluation of construction methods for multi-class problems using binary classifiers

Shigeichi Hirasawa, Gendo Kumoi, Manabu Kobayashi, Masayuki Goto, Hiroshige Inazumi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Construction methods for multi-valued classification (multi-class) systems using binary classifiers are discussed and evaluated by a trade-off model for system evaluation based on rate-distortion theory. Suppose the multi-class systems consisted of M(≥3) categories and N(≥M-1) binary classifiers, then they can be represented by a matrix W, where the matrix W is given by a table of M code words with length N, called a code word table. For a document classification task, the relationship between the probability of classification error Pe and the number of binary classifiers N for given M is investigated, and we show that our constructed systems satisfy desirable properties such as “Flexible”, and “Elastic”. In particular, modified Reed Muller codes perform well: they are shown to be “Effective elastic”. As a second application we consider a hand-written character recognition task, and we show that the desirable properties are also satisfied.

Original languageEnglish
Title of host publicationTrends and Advances in Information Systems and Technologies
PublisherSpringer-Verlag
Pages909-919
Number of pages11
ISBN (Print)9783319777115
DOIs
Publication statusPublished - 2018 Jan 1
Event6th World Conference on Information Systems and Technologies, WorldCIST 2018 - Naples, Italy
Duration: 2018 Mar 272018 Mar 29

Publication series

NameAdvances in Intelligent Systems and Computing
Volume746
ISSN (Print)2194-5357

Other

Other6th World Conference on Information Systems and Technologies, WorldCIST 2018
CountryItaly
CityNaples
Period18/3/2718/3/29

Fingerprint

Classifiers
Character recognition

Keywords

  • Binary classifier
  • ECOC
  • Error correcting codes
  • Exhaustive code
  • Multi-valued classification
  • Trade-off model

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science(all)

Cite this

Hirasawa, S., Kumoi, G., Kobayashi, M., Goto, M., & Inazumi, H. (2018). System evaluation of construction methods for multi-class problems using binary classifiers. In Trends and Advances in Information Systems and Technologies (pp. 909-919). (Advances in Intelligent Systems and Computing; Vol. 746). Springer-Verlag. https://doi.org/10.1007/978-3-319-77712-2_86

System evaluation of construction methods for multi-class problems using binary classifiers. / Hirasawa, Shigeichi; Kumoi, Gendo; Kobayashi, Manabu; Goto, Masayuki; Inazumi, Hiroshige.

Trends and Advances in Information Systems and Technologies. Springer-Verlag, 2018. p. 909-919 (Advances in Intelligent Systems and Computing; Vol. 746).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hirasawa, S, Kumoi, G, Kobayashi, M, Goto, M & Inazumi, H 2018, System evaluation of construction methods for multi-class problems using binary classifiers. in Trends and Advances in Information Systems and Technologies. Advances in Intelligent Systems and Computing, vol. 746, Springer-Verlag, pp. 909-919, 6th World Conference on Information Systems and Technologies, WorldCIST 2018, Naples, Italy, 18/3/27. https://doi.org/10.1007/978-3-319-77712-2_86
Hirasawa S, Kumoi G, Kobayashi M, Goto M, Inazumi H. System evaluation of construction methods for multi-class problems using binary classifiers. In Trends and Advances in Information Systems and Technologies. Springer-Verlag. 2018. p. 909-919. (Advances in Intelligent Systems and Computing). https://doi.org/10.1007/978-3-319-77712-2_86
Hirasawa, Shigeichi ; Kumoi, Gendo ; Kobayashi, Manabu ; Goto, Masayuki ; Inazumi, Hiroshige. / System evaluation of construction methods for multi-class problems using binary classifiers. Trends and Advances in Information Systems and Technologies. Springer-Verlag, 2018. pp. 909-919 (Advances in Intelligent Systems and Computing).
@inproceedings{f70eff4cf7564c6a8bbd8a50ce88da80,
title = "System evaluation of construction methods for multi-class problems using binary classifiers",
abstract = "Construction methods for multi-valued classification (multi-class) systems using binary classifiers are discussed and evaluated by a trade-off model for system evaluation based on rate-distortion theory. Suppose the multi-class systems consisted of M(≥3) categories and N(≥M-1) binary classifiers, then they can be represented by a matrix W, where the matrix W is given by a table of M code words with length N, called a code word table. For a document classification task, the relationship between the probability of classification error Pe and the number of binary classifiers N for given M is investigated, and we show that our constructed systems satisfy desirable properties such as “Flexible”, and “Elastic”. In particular, modified Reed Muller codes perform well: they are shown to be “Effective elastic”. As a second application we consider a hand-written character recognition task, and we show that the desirable properties are also satisfied.",
keywords = "Binary classifier, ECOC, Error correcting codes, Exhaustive code, Multi-valued classification, Trade-off model",
author = "Shigeichi Hirasawa and Gendo Kumoi and Manabu Kobayashi and Masayuki Goto and Hiroshige Inazumi",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-319-77712-2_86",
language = "English",
isbn = "9783319777115",
series = "Advances in Intelligent Systems and Computing",
publisher = "Springer-Verlag",
pages = "909--919",
booktitle = "Trends and Advances in Information Systems and Technologies",

}

TY - GEN

T1 - System evaluation of construction methods for multi-class problems using binary classifiers

AU - Hirasawa, Shigeichi

AU - Kumoi, Gendo

AU - Kobayashi, Manabu

AU - Goto, Masayuki

AU - Inazumi, Hiroshige

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Construction methods for multi-valued classification (multi-class) systems using binary classifiers are discussed and evaluated by a trade-off model for system evaluation based on rate-distortion theory. Suppose the multi-class systems consisted of M(≥3) categories and N(≥M-1) binary classifiers, then they can be represented by a matrix W, where the matrix W is given by a table of M code words with length N, called a code word table. For a document classification task, the relationship between the probability of classification error Pe and the number of binary classifiers N for given M is investigated, and we show that our constructed systems satisfy desirable properties such as “Flexible”, and “Elastic”. In particular, modified Reed Muller codes perform well: they are shown to be “Effective elastic”. As a second application we consider a hand-written character recognition task, and we show that the desirable properties are also satisfied.

AB - Construction methods for multi-valued classification (multi-class) systems using binary classifiers are discussed and evaluated by a trade-off model for system evaluation based on rate-distortion theory. Suppose the multi-class systems consisted of M(≥3) categories and N(≥M-1) binary classifiers, then they can be represented by a matrix W, where the matrix W is given by a table of M code words with length N, called a code word table. For a document classification task, the relationship between the probability of classification error Pe and the number of binary classifiers N for given M is investigated, and we show that our constructed systems satisfy desirable properties such as “Flexible”, and “Elastic”. In particular, modified Reed Muller codes perform well: they are shown to be “Effective elastic”. As a second application we consider a hand-written character recognition task, and we show that the desirable properties are also satisfied.

KW - Binary classifier

KW - ECOC

KW - Error correcting codes

KW - Exhaustive code

KW - Multi-valued classification

KW - Trade-off model

UR - http://www.scopus.com/inward/record.url?scp=85045309941&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045309941&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-77712-2_86

DO - 10.1007/978-3-319-77712-2_86

M3 - Conference contribution

SN - 9783319777115

T3 - Advances in Intelligent Systems and Computing

SP - 909

EP - 919

BT - Trends and Advances in Information Systems and Technologies

PB - Springer-Verlag

ER -