Designing a classifier by a layered multi-population genetic programming approach

Jung Yi Lin, Hao-Ren Ke, Been Chian Chien, Wei Pang Yang

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

This paper proposes a method called layered genetic programming (LAGEP) to construct a classifier based on multi-population genetic programming (MGP). LAGEP employs layer architecture to arrange multiple populations. A layer is composed of a number of populations. The results of populations are discriminant functions. These functions transform the training set to construct a new training set. The successive layer uses the new training set to obtain better discriminant functions. Moreover, because the functions generated by each layer will be composed to a long discriminant function, which is the result of LAGEP, every layer can evolve with short individuals. For each population, we propose an adaptive mutation rate tuning method to increase the mutation rate based on fitness values and remaining generations. Several experiments are conducted with different settings of LAGEP and several real-world medical problems. Experiment results show that LAGEP achieves comparable accuracy to single population GP in much less time.

Original languageEnglish
Pages (from-to)2211-2225
Number of pages15
JournalPattern Recognition
Volume40
Issue number8
DOIs
Publication statusPublished - 2007 Aug 1

Fingerprint

Genetic programming
Classifiers
Medical problems
Population Genetics
Tuning
Experiments

Keywords

  • Classification
  • Evolutionary computation
  • Multi-population genetic programming

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Designing a classifier by a layered multi-population genetic programming approach. / Lin, Jung Yi; Ke, Hao-Ren; Chien, Been Chian; Yang, Wei Pang.

In: Pattern Recognition, Vol. 40, No. 8, 01.08.2007, p. 2211-2225.

Research output: Contribution to journalArticle

Lin, Jung Yi ; Ke, Hao-Ren ; Chien, Been Chian ; Yang, Wei Pang. / Designing a classifier by a layered multi-population genetic programming approach. In: Pattern Recognition. 2007 ; Vol. 40, No. 8. pp. 2211-2225.
@article{6317b988d070407685e43f5ca287d27b,
title = "Designing a classifier by a layered multi-population genetic programming approach",
abstract = "This paper proposes a method called layered genetic programming (LAGEP) to construct a classifier based on multi-population genetic programming (MGP). LAGEP employs layer architecture to arrange multiple populations. A layer is composed of a number of populations. The results of populations are discriminant functions. These functions transform the training set to construct a new training set. The successive layer uses the new training set to obtain better discriminant functions. Moreover, because the functions generated by each layer will be composed to a long discriminant function, which is the result of LAGEP, every layer can evolve with short individuals. For each population, we propose an adaptive mutation rate tuning method to increase the mutation rate based on fitness values and remaining generations. Several experiments are conducted with different settings of LAGEP and several real-world medical problems. Experiment results show that LAGEP achieves comparable accuracy to single population GP in much less time.",
keywords = "Classification, Evolutionary computation, Multi-population genetic programming",
author = "Lin, {Jung Yi} and Hao-Ren Ke and Chien, {Been Chian} and Yang, {Wei Pang}",
year = "2007",
month = "8",
day = "1",
doi = "10.1016/j.patcog.2007.01.003",
language = "English",
volume = "40",
pages = "2211--2225",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "8",

}

TY - JOUR

T1 - Designing a classifier by a layered multi-population genetic programming approach

AU - Lin, Jung Yi

AU - Ke, Hao-Ren

AU - Chien, Been Chian

AU - Yang, Wei Pang

PY - 2007/8/1

Y1 - 2007/8/1

N2 - This paper proposes a method called layered genetic programming (LAGEP) to construct a classifier based on multi-population genetic programming (MGP). LAGEP employs layer architecture to arrange multiple populations. A layer is composed of a number of populations. The results of populations are discriminant functions. These functions transform the training set to construct a new training set. The successive layer uses the new training set to obtain better discriminant functions. Moreover, because the functions generated by each layer will be composed to a long discriminant function, which is the result of LAGEP, every layer can evolve with short individuals. For each population, we propose an adaptive mutation rate tuning method to increase the mutation rate based on fitness values and remaining generations. Several experiments are conducted with different settings of LAGEP and several real-world medical problems. Experiment results show that LAGEP achieves comparable accuracy to single population GP in much less time.

AB - This paper proposes a method called layered genetic programming (LAGEP) to construct a classifier based on multi-population genetic programming (MGP). LAGEP employs layer architecture to arrange multiple populations. A layer is composed of a number of populations. The results of populations are discriminant functions. These functions transform the training set to construct a new training set. The successive layer uses the new training set to obtain better discriminant functions. Moreover, because the functions generated by each layer will be composed to a long discriminant function, which is the result of LAGEP, every layer can evolve with short individuals. For each population, we propose an adaptive mutation rate tuning method to increase the mutation rate based on fitness values and remaining generations. Several experiments are conducted with different settings of LAGEP and several real-world medical problems. Experiment results show that LAGEP achieves comparable accuracy to single population GP in much less time.

KW - Classification

KW - Evolutionary computation

KW - Multi-population genetic programming

UR - http://www.scopus.com/inward/record.url?scp=34247119090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34247119090&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2007.01.003

DO - 10.1016/j.patcog.2007.01.003

M3 - Article

AN - SCOPUS:34247119090

VL - 40

SP - 2211

EP - 2225

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 8

ER -