Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules

Beatriz de la Iglesia; Alan Reynolds; Vic J. Rayward-Smith

doi:10.1007/978-3-540-31880-4_57

Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules

Beatriz de la Iglesia, Alan Reynolds, Vic J. Rayward-Smith

Research output: Chapter in Book/Report/Conference proceeding › Chapter

35 Citations (Scopus)

Abstract

In this paper, we experiment with a combination of innovative approaches to rule induction to encourage the production of interesting sets of classification rules. These include multi-objective metaheuristics to induce the rules; measures of rule dissimilarity to encourage the production of dissimilar rules; and rule clustering algorithms to evaluate the results obtained. Our previous implementation of NSGA-II for rule induction produces a set of cc-optimal rules (coverage-confidence optimal rules). Among the set of rules produced there may be rules that are very similar. We explore the concept of rule similarity and experiment with a number of modifications of the crowding distance to increasing the diversity of the partial classification rules produced by the multi-objective algorithm.

Original language	English
Title of host publication	Evolutionary Multi-Criterion Optimization
Editors	Carlos Coello Coello, Arturo Hernández Aguirre, Eckart Zitzler
Publisher	Springer Berlin / Heidelberg
Pages	826-840
Number of pages	15
Volume	3410
ISBN (Print)	978-3-540-24983-2
DOIs	https://doi.org/10.1007/978-3-540-31880-4_57
Publication status	Published - 2005

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer Berlin / Heidelberg

Access to Document

10.1007/978-3-540-31880-4_57

Cite this

de la Iglesia, B., Reynolds, A., & Rayward-Smith, V. J. (2005). Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules. In C. Coello Coello, A. Hernández Aguirre, & E. Zitzler (Eds.), Evolutionary Multi-Criterion Optimization (Vol. 3410, pp. 826-840). (Lecture Notes in Computer Science). Springer Berlin / Heidelberg. https://doi.org/10.1007/978-3-540-31880-4_57

de la Iglesia, Beatriz ; Reynolds, Alan ; Rayward-Smith, Vic J. / Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules. Evolutionary Multi-Criterion Optimization. editor / Carlos Coello Coello ; Arturo Hernández Aguirre ; Eckart Zitzler. Vol. 3410 Springer Berlin / Heidelberg, 2005. pp. 826-840 (Lecture Notes in Computer Science).

@inbook{25c70f0b029c4b3f9470e57a65914f6d,

title = "Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules",

abstract = "In this paper, we experiment with a combination of innovative approaches to rule induction to encourage the production of interesting sets of classification rules. These include multi-objective metaheuristics to induce the rules; measures of rule dissimilarity to encourage the production of dissimilar rules; and rule clustering algorithms to evaluate the results obtained. Our previous implementation of NSGA-II for rule induction produces a set of cc-optimal rules (coverage-confidence optimal rules). Among the set of rules produced there may be rules that are very similar. We explore the concept of rule similarity and experiment with a number of modifications of the crowding distance to increasing the diversity of the partial classification rules produced by the multi-objective algorithm.",

author = "{de la Iglesia}, Beatriz and Alan Reynolds and Rayward-Smith, {Vic J.}",

note = "Third International Conference, EMO 2005, Guanajuato, Mexico, March 9-11, 2005",

year = "2005",

doi = "10.1007/978-3-540-31880-4_57",

language = "English",

isbn = "978-3-540-24983-2",

volume = "3410",

series = "Lecture Notes in Computer Science",

publisher = "Springer Berlin / Heidelberg",

pages = "826--840",

editor = "{Coello Coello}, Carlos and {Hern{\'a}ndez Aguirre}, Arturo and Eckart Zitzler",

booktitle = "Evolutionary Multi-Criterion Optimization",

}

de la Iglesia, B, Reynolds, A & Rayward-Smith, VJ 2005, Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules. in C Coello Coello, A Hernández Aguirre & E Zitzler (eds), Evolutionary Multi-Criterion Optimization. vol. 3410, Lecture Notes in Computer Science, Springer Berlin / Heidelberg, pp. 826-840. https://doi.org/10.1007/978-3-540-31880-4_57

Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules. / de la Iglesia, Beatriz; Reynolds, Alan; Rayward-Smith, Vic J.

Evolutionary Multi-Criterion Optimization. ed. / Carlos Coello Coello; Arturo Hernández Aguirre; Eckart Zitzler. Vol. 3410 Springer Berlin / Heidelberg, 2005. p. 826-840 (Lecture Notes in Computer Science).

Research output: Chapter in Book/Report/Conference proceeding › Chapter

TY - CHAP

T1 - Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules

AU - de la Iglesia, Beatriz

AU - Reynolds, Alan

AU - Rayward-Smith, Vic J.

N1 - Third International Conference, EMO 2005, Guanajuato, Mexico, March 9-11, 2005

PY - 2005

Y1 - 2005

N2 - In this paper, we experiment with a combination of innovative approaches to rule induction to encourage the production of interesting sets of classification rules. These include multi-objective metaheuristics to induce the rules; measures of rule dissimilarity to encourage the production of dissimilar rules; and rule clustering algorithms to evaluate the results obtained. Our previous implementation of NSGA-II for rule induction produces a set of cc-optimal rules (coverage-confidence optimal rules). Among the set of rules produced there may be rules that are very similar. We explore the concept of rule similarity and experiment with a number of modifications of the crowding distance to increasing the diversity of the partial classification rules produced by the multi-objective algorithm.

AB - In this paper, we experiment with a combination of innovative approaches to rule induction to encourage the production of interesting sets of classification rules. These include multi-objective metaheuristics to induce the rules; measures of rule dissimilarity to encourage the production of dissimilar rules; and rule clustering algorithms to evaluate the results obtained. Our previous implementation of NSGA-II for rule induction produces a set of cc-optimal rules (coverage-confidence optimal rules). Among the set of rules produced there may be rules that are very similar. We explore the concept of rule similarity and experiment with a number of modifications of the crowding distance to increasing the diversity of the partial classification rules produced by the multi-objective algorithm.

U2 - 10.1007/978-3-540-31880-4_57

DO - 10.1007/978-3-540-31880-4_57

M3 - Chapter

SN - 978-3-540-24983-2

VL - 3410

T3 - Lecture Notes in Computer Science

SP - 826

EP - 840

BT - Evolutionary Multi-Criterion Optimization

A2 - Coello Coello, Carlos

A2 - Hernández Aguirre, Arturo

A2 - Zitzler, Eckart

PB - Springer Berlin / Heidelberg

ER -

de la Iglesia B, Reynolds A, Rayward-Smith VJ. Developments on a Multi-objective Metaheuristic (MOMH) Algorithm for Finding Interesting Sets of Classification Rules. In Coello Coello C, Hernández Aguirre A, Zitzler E, editors, Evolutionary Multi-Criterion Optimization. Vol. 3410. Springer Berlin / Heidelberg. 2005. p. 826-840. (Lecture Notes in Computer Science). doi: 10.1007/978-3-540-31880-4_57