Exploiting the relationships among several binary classifiers via data transformation

Kar Ann Toh, Geok Choo Tan

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

The structural resemblance among several existing classifiers has motivated us to investigate their underlying relationships. By exploring into the mapping solutions of these classifiers, we found that they can be linked by simple feature data scaling. In other words, the key to these relationships lies upon how the replica of feature data are being scaled. This finding leads us directly to an exploration of novel classifiers beyond existing settings. Based on an extensive empirical evaluation, we show that the proposed formulation facilitates a tuning capability beyond existing settings for classifier generalization.

Original languageEnglish
Pages (from-to)1509-1522
Number of pages14
JournalPattern Recognition
Volume47
Issue number3
DOIs
Publication statusPublished - 2014 Mar 1

Fingerprint

Classifiers
Tuning

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{3fd21dbee92642faab12ea207f911cc7,
title = "Exploiting the relationships among several binary classifiers via data transformation",
abstract = "The structural resemblance among several existing classifiers has motivated us to investigate their underlying relationships. By exploring into the mapping solutions of these classifiers, we found that they can be linked by simple feature data scaling. In other words, the key to these relationships lies upon how the replica of feature data are being scaled. This finding leads us directly to an exploration of novel classifiers beyond existing settings. Based on an extensive empirical evaluation, we show that the proposed formulation facilitates a tuning capability beyond existing settings for classifier generalization.",
author = "Toh, {Kar Ann} and Tan, {Geok Choo}",
year = "2014",
month = "3",
day = "1",
doi = "10.1016/j.patcog.2013.09.030",
language = "English",
volume = "47",
pages = "1509--1522",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "3",

}

Exploiting the relationships among several binary classifiers via data transformation. / Toh, Kar Ann; Tan, Geok Choo.

In: Pattern Recognition, Vol. 47, No. 3, 01.03.2014, p. 1509-1522.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Exploiting the relationships among several binary classifiers via data transformation

AU - Toh, Kar Ann

AU - Tan, Geok Choo

PY - 2014/3/1

Y1 - 2014/3/1

N2 - The structural resemblance among several existing classifiers has motivated us to investigate their underlying relationships. By exploring into the mapping solutions of these classifiers, we found that they can be linked by simple feature data scaling. In other words, the key to these relationships lies upon how the replica of feature data are being scaled. This finding leads us directly to an exploration of novel classifiers beyond existing settings. Based on an extensive empirical evaluation, we show that the proposed formulation facilitates a tuning capability beyond existing settings for classifier generalization.

AB - The structural resemblance among several existing classifiers has motivated us to investigate their underlying relationships. By exploring into the mapping solutions of these classifiers, we found that they can be linked by simple feature data scaling. In other words, the key to these relationships lies upon how the replica of feature data are being scaled. This finding leads us directly to an exploration of novel classifiers beyond existing settings. Based on an extensive empirical evaluation, we show that the proposed formulation facilitates a tuning capability beyond existing settings for classifier generalization.

UR - http://www.scopus.com/inward/record.url?scp=84888341583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888341583&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2013.09.030

DO - 10.1016/j.patcog.2013.09.030

M3 - Article

VL - 47

SP - 1509

EP - 1522

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 3

ER -