Comparisons between heuristics based on correlativity and efficiency for landmarker generation

Daren Ler, Irena Koprinska, Sanjay Chawla

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

1 Citation (Scopus)

Abstract

Recently, we proposed a new meta-learning approach based on landmarking. This approach, which utilises a new set of criteria for selecting landmarkers, generates a set of landmarkers that are each functions over the performance over subsets of the candidate algorithms being landmarked. In this paper, we experiment with three heuristics based on correlativity and efficiency. With each heuristic, the landmarkers generated using linear regression are able to estimate accuracy well, even when only utilising a small fraction of the given algorithms. The results also show that the heuristic in which efficiencies are estimated via 1-nearest neighbour outperformed the other heuristics.

Original languageEnglish
Title of host publicationProceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems
EditorsM. Ishikawa, S. Hashimoto, M. Paprzycki, E. Barakova, K. Yoshida, M. Koppen, D.M. Corne, A. Abraham
Pages32-37
Number of pages6
Publication statusPublished - 2005
Externally publishedYes
EventProceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems - Kitakyushu
Duration: 5 Dec 20048 Dec 2004

Other

OtherProceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems
CityKitakyushu
Period5/12/048/12/04

Fingerprint

Set theory
Linear regression
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ler, D., Koprinska, I., & Chawla, S. (2005). Comparisons between heuristics based on correlativity and efficiency for landmarker generation. In M. Ishikawa, S. Hashimoto, M. Paprzycki, E. Barakova, K. Yoshida, M. Koppen, D. M. Corne, ... A. Abraham (Eds.), Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems (pp. 32-37)

Comparisons between heuristics based on correlativity and efficiency for landmarker generation. / Ler, Daren; Koprinska, Irena; Chawla, Sanjay.

Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems. ed. / M. Ishikawa; S. Hashimoto; M. Paprzycki; E. Barakova; K. Yoshida; M. Koppen; D.M. Corne; A. Abraham. 2005. p. 32-37.

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

Ler, D, Koprinska, I & Chawla, S 2005, Comparisons between heuristics based on correlativity and efficiency for landmarker generation. in M Ishikawa, S Hashimoto, M Paprzycki, E Barakova, K Yoshida, M Koppen, DM Corne & A Abraham (eds), Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems. pp. 32-37, Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems, Kitakyushu, 5/12/04.
Ler D, Koprinska I, Chawla S. Comparisons between heuristics based on correlativity and efficiency for landmarker generation. In Ishikawa M, Hashimoto S, Paprzycki M, Barakova E, Yoshida K, Koppen M, Corne DM, Abraham A, editors, Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems. 2005. p. 32-37
Ler, Daren ; Koprinska, Irena ; Chawla, Sanjay. / Comparisons between heuristics based on correlativity and efficiency for landmarker generation. Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems. editor / M. Ishikawa ; S. Hashimoto ; M. Paprzycki ; E. Barakova ; K. Yoshida ; M. Koppen ; D.M. Corne ; A. Abraham. 2005. pp. 32-37
@inproceedings{e16ae3b9c60443b4a911b7fe48ef76b1,
title = "Comparisons between heuristics based on correlativity and efficiency for landmarker generation",
abstract = "Recently, we proposed a new meta-learning approach based on landmarking. This approach, which utilises a new set of criteria for selecting landmarkers, generates a set of landmarkers that are each functions over the performance over subsets of the candidate algorithms being landmarked. In this paper, we experiment with three heuristics based on correlativity and efficiency. With each heuristic, the landmarkers generated using linear regression are able to estimate accuracy well, even when only utilising a small fraction of the given algorithms. The results also show that the heuristic in which efficiencies are estimated via 1-nearest neighbour outperformed the other heuristics.",
author = "Daren Ler and Irena Koprinska and Sanjay Chawla",
year = "2005",
language = "English",
isbn = "0769522912",
pages = "32--37",
editor = "M. Ishikawa and S. Hashimoto and M. Paprzycki and E. Barakova and K. Yoshida and M. Koppen and D.M. Corne and A. Abraham",
booktitle = "Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems",

}

TY - GEN

T1 - Comparisons between heuristics based on correlativity and efficiency for landmarker generation

AU - Ler, Daren

AU - Koprinska, Irena

AU - Chawla, Sanjay

PY - 2005

Y1 - 2005

N2 - Recently, we proposed a new meta-learning approach based on landmarking. This approach, which utilises a new set of criteria for selecting landmarkers, generates a set of landmarkers that are each functions over the performance over subsets of the candidate algorithms being landmarked. In this paper, we experiment with three heuristics based on correlativity and efficiency. With each heuristic, the landmarkers generated using linear regression are able to estimate accuracy well, even when only utilising a small fraction of the given algorithms. The results also show that the heuristic in which efficiencies are estimated via 1-nearest neighbour outperformed the other heuristics.

AB - Recently, we proposed a new meta-learning approach based on landmarking. This approach, which utilises a new set of criteria for selecting landmarkers, generates a set of landmarkers that are each functions over the performance over subsets of the candidate algorithms being landmarked. In this paper, we experiment with three heuristics based on correlativity and efficiency. With each heuristic, the landmarkers generated using linear regression are able to estimate accuracy well, even when only utilising a small fraction of the given algorithms. The results also show that the heuristic in which efficiencies are estimated via 1-nearest neighbour outperformed the other heuristics.

UR - http://www.scopus.com/inward/record.url?scp=28344439263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=28344439263&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:28344439263

SN - 0769522912

SP - 32

EP - 37

BT - Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems

A2 - Ishikawa, M.

A2 - Hashimoto, S.

A2 - Paprzycki, M.

A2 - Barakova, E.

A2 - Yoshida, K.

A2 - Koppen, M.

A2 - Corne, D.M.

A2 - Abraham, A.

ER -