An improved multi-label classification method and its application to functional genomics

Benhui Chen, Weifeng Gu, Takayuki Furuzuki

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

In this paper, a multi-label classification method based on label ranking and delicate boundary Support Vector Machine (SVM) is proposed for solving the functional genomics applications. Firstly, an improved probabilistic SVM with delicate decision boundary is used as scoring approach to obtain a proper label rank. Secondly, an instance-dependent thresholding strategy is proposed to decide classification results. A d-folds validation approach is utilised to determine a set of target thresholds for all training samples as teachers, then an appropriate instance-dependent threshold for each testing instance is obtained by applying k-Nearest Neighbours (KNN) strategy on this teacher threshold set.

Original languageEnglish
Pages (from-to)133-145
Number of pages13
JournalInternational Journal of Computational Biology and Drug Design
Volume3
Issue number2
DOIs
Publication statusPublished - 2010 Sep

Fingerprint

Genomics
Labels
Support vector machines
Testing
Support Vector Machine

Keywords

  • Functional genomics
  • Multi-label classification
  • Ranking based method
  • Support vector machine
  • SVM
  • Thresholding strategy

ASJC Scopus subject areas

  • Computer Science Applications
  • Drug Discovery

Cite this

An improved multi-label classification method and its application to functional genomics. / Chen, Benhui; Gu, Weifeng; Furuzuki, Takayuki.

In: International Journal of Computational Biology and Drug Design, Vol. 3, No. 2, 09.2010, p. 133-145.

Research output: Contribution to journalArticle

@article{0ddaab135dc04b818eda22212bf95214,
title = "An improved multi-label classification method and its application to functional genomics",
abstract = "In this paper, a multi-label classification method based on label ranking and delicate boundary Support Vector Machine (SVM) is proposed for solving the functional genomics applications. Firstly, an improved probabilistic SVM with delicate decision boundary is used as scoring approach to obtain a proper label rank. Secondly, an instance-dependent thresholding strategy is proposed to decide classification results. A d-folds validation approach is utilised to determine a set of target thresholds for all training samples as teachers, then an appropriate instance-dependent threshold for each testing instance is obtained by applying k-Nearest Neighbours (KNN) strategy on this teacher threshold set.",
keywords = "Functional genomics, Multi-label classification, Ranking based method, Support vector machine, SVM, Thresholding strategy",
author = "Benhui Chen and Weifeng Gu and Takayuki Furuzuki",
year = "2010",
month = "9",
doi = "10.1504/IJCBDD.2010.035239",
language = "English",
volume = "3",
pages = "133--145",
journal = "International Journal of Computational Biology and Drug Design",
issn = "1756-0756",
publisher = "Inderscience Enterprises Ltd",
number = "2",

}

TY - JOUR

T1 - An improved multi-label classification method and its application to functional genomics

AU - Chen, Benhui

AU - Gu, Weifeng

AU - Furuzuki, Takayuki

PY - 2010/9

Y1 - 2010/9

N2 - In this paper, a multi-label classification method based on label ranking and delicate boundary Support Vector Machine (SVM) is proposed for solving the functional genomics applications. Firstly, an improved probabilistic SVM with delicate decision boundary is used as scoring approach to obtain a proper label rank. Secondly, an instance-dependent thresholding strategy is proposed to decide classification results. A d-folds validation approach is utilised to determine a set of target thresholds for all training samples as teachers, then an appropriate instance-dependent threshold for each testing instance is obtained by applying k-Nearest Neighbours (KNN) strategy on this teacher threshold set.

AB - In this paper, a multi-label classification method based on label ranking and delicate boundary Support Vector Machine (SVM) is proposed for solving the functional genomics applications. Firstly, an improved probabilistic SVM with delicate decision boundary is used as scoring approach to obtain a proper label rank. Secondly, an instance-dependent thresholding strategy is proposed to decide classification results. A d-folds validation approach is utilised to determine a set of target thresholds for all training samples as teachers, then an appropriate instance-dependent threshold for each testing instance is obtained by applying k-Nearest Neighbours (KNN) strategy on this teacher threshold set.

KW - Functional genomics

KW - Multi-label classification

KW - Ranking based method

KW - Support vector machine

KW - SVM

KW - Thresholding strategy

UR - http://www.scopus.com/inward/record.url?scp=77956979399&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956979399&partnerID=8YFLogxK

U2 - 10.1504/IJCBDD.2010.035239

DO - 10.1504/IJCBDD.2010.035239

M3 - Article

C2 - 20852337

AN - SCOPUS:77956979399

VL - 3

SP - 133

EP - 145

JO - International Journal of Computational Biology and Drug Design

JF - International Journal of Computational Biology and Drug Design

SN - 1756-0756

IS - 2

ER -