Genetic approach to constructive induction based on non-algebraic feature representation

Leila S. Shafti; Eduardo Pérez

doi:10.1007/978-3-540-45231-7_55

Genetic approach to constructive induction based on non-algebraic feature representation

Leila S. Shafti^*
, Eduardo Pérez

^*Corresponding author for this work

Computer And Information Sciences

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

12 Citations (Scopus)

Abstract

The aim of constructive induction (CI) is to transform the original data representation of hard concepts with complex interaction into one that outlines the relation among attributes. CI methods based on greedy search suffer from the local optima problem because of high variation in the search space of hard learning problems. To reduce the local optima problem, we propose a CI method based on genetic (evolutionary) algorithms. The method comprises two integrated genetic algorithms to construct functions over subsets of attributes in order to highlight regularities for the learner. Using non-algebraic representation for constructed functions assigns an equal degree of complexity to functions. This reduces the difficulty of constructing complex features. Experiments show that our method is comparable with and in some cases superior to existing CI methods.

Original language	English
Title of host publication	Advances in Intelligent Data Analysis V
Subtitle of host publication	5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings
Editors	Michael R. Berthold, Hans-Joachim Lenz, Elizabeth Bradley, Rudolf Kruse, Christian Borgelt
Publisher	Springer
Pages	599-610
Number of pages	12
ISBN (Electronic)	9783540452317
ISBN (Print)	9783540408130
DOIs	https://doi.org/10.1007/978-3-540-45231-7_55
Publication status	Published - 2003
Event	5th International Symposium on Intelligent Data Analysis - Berlin, Germany Duration: 28 Aug 2003 → 30 Aug 2003

Publication series

Name	Lecture Notes in Computer Science
Volume	2810
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Symposium on Intelligent Data Analysis
Abbreviated title	IDA 2003
Country/Territory	Germany
City	Berlin
Period	28/08/03 → 30/08/03

Funding

This work has been partially supported by the Spanish Interdepartmental Commission for Science and Technology (CICYT), under Grants numbers TIC98-0247-C02-02 and TIC2002-1948.

Keywords

constructive induction
local optima problem
non-algebraic representation
genetic algorithms

Access to Document

10.1007/978-3-540-45231-7_55

Cite this

Shafti, L. S., & Pérez, E. (2003). Genetic approach to constructive induction based on non-algebraic feature representation. In M. R. Berthold, H.-J. Lenz, E. Bradley, R. Kruse, & C. Borgelt (Eds.), Advances in Intelligent Data Analysis V: 5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings (pp. 599-610). (Lecture Notes in Computer Science; Vol. 2810). Springer. https://doi.org/10.1007/978-3-540-45231-7_55

Shafti, Leila S. ; Pérez, Eduardo. / Genetic approach to constructive induction based on non-algebraic feature representation. Advances in Intelligent Data Analysis V: 5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings. editor / Michael R. Berthold ; Hans-Joachim Lenz ; Elizabeth Bradley ; Rudolf Kruse ; Christian Borgelt. Springer, 2003. pp. 599-610 (Lecture Notes in Computer Science).

@inproceedings{169770a475224ce284cd94e24544ebe0,

title = "Genetic approach to constructive induction based on non-algebraic feature representation",

abstract = "The aim of constructive induction (CI) is to transform the original data representation of hard concepts with complex interaction into one that outlines the relation among attributes. CI methods based on greedy search suffer from the local optima problem because of high variation in the search space of hard learning problems. To reduce the local optima problem, we propose a CI method based on genetic (evolutionary) algorithms. The method comprises two integrated genetic algorithms to construct functions over subsets of attributes in order to highlight regularities for the learner. Using non-algebraic representation for constructed functions assigns an equal degree of complexity to functions. This reduces the difficulty of constructing complex features. Experiments show that our method is comparable with and in some cases superior to existing CI methods.",

keywords = "constructive induction, local optima problem, non-algebraic representation, genetic algorithms",

author = "Shafti, \{Leila S.\} and Eduardo P{\'e}rez",

year = "2003",

doi = "10.1007/978-3-540-45231-7\_55",

language = "English",

isbn = "9783540408130",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "599--610",

editor = "Berthold, \{Michael R.\} and Hans-Joachim Lenz and Elizabeth Bradley and Rudolf Kruse and Christian Borgelt",

booktitle = "Advances in Intelligent Data Analysis V",

address = "Switzerland",

note = "5th International Symposium on Intelligent Data Analysis, IDA 2003 ; Conference date: 28-08-2003 Through 30-08-2003",

}

Shafti, LS & Pérez, E 2003, Genetic approach to constructive induction based on non-algebraic feature representation. in MR Berthold, H-J Lenz, E Bradley, R Kruse & C Borgelt (eds), Advances in Intelligent Data Analysis V: 5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings. Lecture Notes in Computer Science, vol. 2810, Springer, pp. 599-610, 5th International Symposium on Intelligent Data Analysis, Berlin, Germany, 28/08/03. https://doi.org/10.1007/978-3-540-45231-7_55

Genetic approach to constructive induction based on non-algebraic feature representation. / Shafti, Leila S.; Pérez, Eduardo.
Advances in Intelligent Data Analysis V: 5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings. ed. / Michael R. Berthold; Hans-Joachim Lenz; Elizabeth Bradley; Rudolf Kruse; Christian Borgelt. Springer, 2003. p. 599-610 (Lecture Notes in Computer Science; Vol. 2810).

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

TY - GEN

T1 - Genetic approach to constructive induction based on non-algebraic feature representation

AU - Shafti, Leila S.

AU - Pérez, Eduardo

PY - 2003

Y1 - 2003

N2 - The aim of constructive induction (CI) is to transform the original data representation of hard concepts with complex interaction into one that outlines the relation among attributes. CI methods based on greedy search suffer from the local optima problem because of high variation in the search space of hard learning problems. To reduce the local optima problem, we propose a CI method based on genetic (evolutionary) algorithms. The method comprises two integrated genetic algorithms to construct functions over subsets of attributes in order to highlight regularities for the learner. Using non-algebraic representation for constructed functions assigns an equal degree of complexity to functions. This reduces the difficulty of constructing complex features. Experiments show that our method is comparable with and in some cases superior to existing CI methods.

AB - The aim of constructive induction (CI) is to transform the original data representation of hard concepts with complex interaction into one that outlines the relation among attributes. CI methods based on greedy search suffer from the local optima problem because of high variation in the search space of hard learning problems. To reduce the local optima problem, we propose a CI method based on genetic (evolutionary) algorithms. The method comprises two integrated genetic algorithms to construct functions over subsets of attributes in order to highlight regularities for the learner. Using non-algebraic representation for constructed functions assigns an equal degree of complexity to functions. This reduces the difficulty of constructing complex features. Experiments show that our method is comparable with and in some cases superior to existing CI methods.

KW - constructive induction

KW - local optima problem

KW - non-algebraic representation

KW - genetic algorithms

U2 - 10.1007/978-3-540-45231-7_55

DO - 10.1007/978-3-540-45231-7_55

M3 - Conference contribution book

AN - SCOPUS:35248833168

SN - 9783540408130

T3 - Lecture Notes in Computer Science

SP - 599

EP - 610

BT - Advances in Intelligent Data Analysis V

A2 - Berthold, Michael R.

A2 - Lenz, Hans-Joachim

A2 - Bradley, Elizabeth

A2 - Kruse, Rudolf

A2 - Borgelt, Christian

PB - Springer

T2 - 5th International Symposium on Intelligent Data Analysis

Y2 - 28 August 2003 through 30 August 2003

ER -

Shafti LS, Pérez E. Genetic approach to constructive induction based on non-algebraic feature representation. In Berthold MR, Lenz HJ, Bradley E, Kruse R, Borgelt C, editors, Advances in Intelligent Data Analysis V: 5th International Symposium on Intelligent Data Analysis, IDA 2003 Berlin, Germany, August 28-30, 2003 Proceedings. Springer. 2003. p. 599-610. (Lecture Notes in Computer Science). doi: 10.1007/978-3-540-45231-7_55

Genetic approach to constructive induction based on non-algebraic feature representation

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Fingerprint

Cite this