A new landmarker generation algorithm based on correlativity

Daren Ler, Irena Koprinska, Sanjay Chawla

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

4 Citations (Scopus)

Abstract

Landmarking is a recent and promising metalearning strategy, which defines meta-features that are themselves efficient learning algorithms. However, the choice of landmarkers is made in an ad hoc manner. In this paper, we propose a new perspective and set of criteria for landmarkers. With these, we introduce a landmarker generation algorithm, which creates a set of landmarkers that each utilise subsets of the algorithms being landmarked. The experiments show that the landmarkers formed, when used with linear regression, are able to estimate accuracy well, even when utilising a small fraction of the given algorithms.

Original languageEnglish
Title of host publicationProceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04
EditorsM. Kantardzic, O. Nasraoui, M. Milanova
Pages178-185
Number of pages8
Publication statusPublished - 2004
Externally publishedYes
Event2004 International Conference on Machine Learning and Applications, ICMLA '04 - Louisville, KY
Duration: 16 Dec 200418 Dec 2004

Other

Other2004 International Conference on Machine Learning and Applications, ICMLA '04
CityLouisville, KY
Period16/12/0418/12/04

Fingerprint

Set theory
Linear regression
Learning algorithms
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ler, D., Koprinska, I., & Chawla, S. (2004). A new landmarker generation algorithm based on correlativity. In M. Kantardzic, O. Nasraoui, & M. Milanova (Eds.), Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04 (pp. 178-185)

A new landmarker generation algorithm based on correlativity. / Ler, Daren; Koprinska, Irena; Chawla, Sanjay.

Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04. ed. / M. Kantardzic; O. Nasraoui; M. Milanova. 2004. p. 178-185.

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

Ler, D, Koprinska, I & Chawla, S 2004, A new landmarker generation algorithm based on correlativity. in M Kantardzic, O Nasraoui & M Milanova (eds), Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04. pp. 178-185, 2004 International Conference on Machine Learning and Applications, ICMLA '04, Louisville, KY, 16/12/04.
Ler D, Koprinska I, Chawla S. A new landmarker generation algorithm based on correlativity. In Kantardzic M, Nasraoui O, Milanova M, editors, Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04. 2004. p. 178-185
Ler, Daren ; Koprinska, Irena ; Chawla, Sanjay. / A new landmarker generation algorithm based on correlativity. Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04. editor / M. Kantardzic ; O. Nasraoui ; M. Milanova. 2004. pp. 178-185
@inproceedings{a5717d5a430346d2bcc169ac3d601dd4,
title = "A new landmarker generation algorithm based on correlativity",
abstract = "Landmarking is a recent and promising metalearning strategy, which defines meta-features that are themselves efficient learning algorithms. However, the choice of landmarkers is made in an ad hoc manner. In this paper, we propose a new perspective and set of criteria for landmarkers. With these, we introduce a landmarker generation algorithm, which creates a set of landmarkers that each utilise subsets of the algorithms being landmarked. The experiments show that the landmarkers formed, when used with linear regression, are able to estimate accuracy well, even when utilising a small fraction of the given algorithms.",
author = "Daren Ler and Irena Koprinska and Sanjay Chawla",
year = "2004",
language = "English",
isbn = "0780388232",
pages = "178--185",
editor = "M. Kantardzic and O. Nasraoui and M. Milanova",
booktitle = "Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04",

}

TY - GEN

T1 - A new landmarker generation algorithm based on correlativity

AU - Ler, Daren

AU - Koprinska, Irena

AU - Chawla, Sanjay

PY - 2004

Y1 - 2004

N2 - Landmarking is a recent and promising metalearning strategy, which defines meta-features that are themselves efficient learning algorithms. However, the choice of landmarkers is made in an ad hoc manner. In this paper, we propose a new perspective and set of criteria for landmarkers. With these, we introduce a landmarker generation algorithm, which creates a set of landmarkers that each utilise subsets of the algorithms being landmarked. The experiments show that the landmarkers formed, when used with linear regression, are able to estimate accuracy well, even when utilising a small fraction of the given algorithms.

AB - Landmarking is a recent and promising metalearning strategy, which defines meta-features that are themselves efficient learning algorithms. However, the choice of landmarkers is made in an ad hoc manner. In this paper, we propose a new perspective and set of criteria for landmarkers. With these, we introduce a landmarker generation algorithm, which creates a set of landmarkers that each utilise subsets of the algorithms being landmarked. The experiments show that the landmarkers formed, when used with linear regression, are able to estimate accuracy well, even when utilising a small fraction of the given algorithms.

UR - http://www.scopus.com/inward/record.url?scp=21244449557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21244449557&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780388232

SP - 178

EP - 185

BT - Proceedings of the 2004 International Conference on Machine Learning and Applications, ICMLA '04

A2 - Kantardzic, M.

A2 - Nasraoui, O.

A2 - Milanova, M.

ER -