Data reduction by genetic algorithms and non-algebraic feature construction: a case study

Leila S. Shafti; Eduardo Pérez

doi:10.1109/HIS.2008.114

Data reduction by genetic algorithms and non-algebraic feature construction: a case study

Leila S. Shafti, Eduardo Pérez

Computer And Information Sciences

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

5 Citations (Scopus)

Abstract

Real-world data are often prepared for purposes other than data mining and machine learning and, therefore, are represented by primitive attributes. When data representation is primitive, preprocessing data before looking for patterns becomes necessary. If lack of domain experts prevents the use of highly informative attributes, patterns are hard to uncover due to complex attribute interactions. This article suggests a new use of MFE3/GA to restructure the primitive data representation by means of capturing and compacting hidden information into new features in order to highlight them to the learner. Empirical results on Poker Hand data set show that the new use successfully improves learning this concept by means of data reduction, generation of a smaller decision tree classifier, and accuracy improvement.

Original language	English
Title of host publication	2008 Eighth International Conference on Hybrid Intelligent Systems
Editors	Fatos Xhafa, Francisco Herrera, Ajith Abraham, Mario Köppen, Jose Manuel Bénitez
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	573-578
Number of pages	6
ISBN (Print)	9780769533261
DOIs	https://doi.org/10.1109/HIS.2008.114
Publication status	Published - 19 Sept 2008
Event	Eighth International Conference on Hybrid Intelligent Systems - Barcelona, Spain Duration: 10 Sept 2008 → 12 Sept 2008

Publication series

Name	International Conference on Hybrid Intelligent Systems (HIS)

Conference

Conference	Eighth International Conference on Hybrid Intelligent Systems
Abbreviated title	HIS 2008
Country/Territory	Spain
City	Barcelona
Period	10/09/08 → 12/09/08

Funding

This work has been partially supported by the Spanish Ministry of Science and Technology, under Grant number TSI2005-08225-C07-06.

Keywords

genetic algorithms
data reduction
MFE3/GA

Access to Document

10.1109/HIS.2008.114

Cite this

Shafti, L. S., & Pérez, E. (2008). Data reduction by genetic algorithms and non-algebraic feature construction: a case study. In F. Xhafa, F. Herrera, A. Abraham, M. Köppen, & J. M. Bénitez (Eds.), 2008 Eighth International Conference on Hybrid Intelligent Systems (pp. 573-578). (International Conference on Hybrid Intelligent Systems (HIS)). IEEE. https://doi.org/10.1109/HIS.2008.114

Shafti, Leila S. ; Pérez, Eduardo. / Data reduction by genetic algorithms and non-algebraic feature construction : a case study. 2008 Eighth International Conference on Hybrid Intelligent Systems. editor / Fatos Xhafa ; Francisco Herrera ; Ajith Abraham ; Mario Köppen ; Jose Manuel Bénitez. Piscataway, NJ : IEEE, 2008. pp. 573-578 (International Conference on Hybrid Intelligent Systems (HIS)).

@inproceedings{5fc543f41d1e40f89f37e0508989981c,

title = "Data reduction by genetic algorithms and non-algebraic feature construction: a case study",

abstract = "Real-world data are often prepared for purposes other than data mining and machine learning and, therefore, are represented by primitive attributes. When data representation is primitive, preprocessing data before looking for patterns becomes necessary. If lack of domain experts prevents the use of highly informative attributes, patterns are hard to uncover due to complex attribute interactions. This article suggests a new use of MFE3/GA to restructure the primitive data representation by means of capturing and compacting hidden information into new features in order to highlight them to the learner. Empirical results on Poker Hand data set show that the new use successfully improves learning this concept by means of data reduction, generation of a smaller decision tree classifier, and accuracy improvement.",

keywords = "genetic algorithms, data reduction, MFE3/GA",

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

year = "2008",

month = sep,

day = "19",

doi = "10.1109/HIS.2008.114",

language = "English",

isbn = "9780769533261 ",

series = "International Conference on Hybrid Intelligent Systems (HIS)",

publisher = "IEEE",

pages = "573--578",

editor = "Fatos Xhafa and Francisco Herrera and Ajith Abraham and Mario K{\"o}ppen and B{\'e}nitez, {Jose Manuel}",

booktitle = "2008 Eighth International Conference on Hybrid Intelligent Systems",

note = "Eighth International Conference on Hybrid Intelligent Systems, HIS 2008 ; Conference date: 10-09-2008 Through 12-09-2008",

}

Shafti, LS & Pérez, E 2008, Data reduction by genetic algorithms and non-algebraic feature construction: a case study. in F Xhafa, F Herrera, A Abraham, M Köppen & JM Bénitez (eds), 2008 Eighth International Conference on Hybrid Intelligent Systems. International Conference on Hybrid Intelligent Systems (HIS), IEEE, Piscataway, NJ, pp. 573-578, Eighth International Conference on Hybrid Intelligent Systems, Barcelona, Spain, 10/09/08. https://doi.org/10.1109/HIS.2008.114

Data reduction by genetic algorithms and non-algebraic feature construction: a case study. / Shafti, Leila S.; Pérez, Eduardo.
2008 Eighth International Conference on Hybrid Intelligent Systems. ed. / Fatos Xhafa; Francisco Herrera; Ajith Abraham; Mario Köppen; Jose Manuel Bénitez. Piscataway, NJ: IEEE, 2008. p. 573-578 (International Conference on Hybrid Intelligent Systems (HIS)).

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

TY - GEN

T1 - Data reduction by genetic algorithms and non-algebraic feature construction

T2 - Eighth International Conference on Hybrid Intelligent Systems

AU - Shafti, Leila S.

AU - Pérez, Eduardo

PY - 2008/9/19

Y1 - 2008/9/19

N2 - Real-world data are often prepared for purposes other than data mining and machine learning and, therefore, are represented by primitive attributes. When data representation is primitive, preprocessing data before looking for patterns becomes necessary. If lack of domain experts prevents the use of highly informative attributes, patterns are hard to uncover due to complex attribute interactions. This article suggests a new use of MFE3/GA to restructure the primitive data representation by means of capturing and compacting hidden information into new features in order to highlight them to the learner. Empirical results on Poker Hand data set show that the new use successfully improves learning this concept by means of data reduction, generation of a smaller decision tree classifier, and accuracy improvement.

AB - Real-world data are often prepared for purposes other than data mining and machine learning and, therefore, are represented by primitive attributes. When data representation is primitive, preprocessing data before looking for patterns becomes necessary. If lack of domain experts prevents the use of highly informative attributes, patterns are hard to uncover due to complex attribute interactions. This article suggests a new use of MFE3/GA to restructure the primitive data representation by means of capturing and compacting hidden information into new features in order to highlight them to the learner. Empirical results on Poker Hand data set show that the new use successfully improves learning this concept by means of data reduction, generation of a smaller decision tree classifier, and accuracy improvement.

KW - genetic algorithms

KW - data reduction

KW - MFE3/GA

U2 - 10.1109/HIS.2008.114

DO - 10.1109/HIS.2008.114

M3 - Conference contribution book

AN - SCOPUS:55349113637

SN - 9780769533261

T3 - International Conference on Hybrid Intelligent Systems (HIS)

SP - 573

EP - 578

BT - 2008 Eighth International Conference on Hybrid Intelligent Systems

A2 - Xhafa, Fatos

A2 - Herrera, Francisco

A2 - Abraham, Ajith

A2 - Köppen, Mario

A2 - Bénitez, Jose Manuel

PB - IEEE

CY - Piscataway, NJ

Y2 - 10 September 2008 through 12 September 2008

ER -

Shafti LS, Pérez E. Data reduction by genetic algorithms and non-algebraic feature construction: a case study. In Xhafa F, Herrera F, Abraham A, Köppen M, Bénitez JM, editors, 2008 Eighth International Conference on Hybrid Intelligent Systems. Piscataway, NJ: IEEE. 2008. p. 573-578. (International Conference on Hybrid Intelligent Systems (HIS)). doi: 10.1109/HIS.2008.114

Data reduction by genetic algorithms and non-algebraic feature construction: a case study

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Fingerprint

Cite this