Recent developments on evolutionary computation techniques to feature construction

Idheba Mohamad Ali O. Swesi, Azuraliza Abu Bakar

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The quality of the search space is an important factor that influences the performance of any machine learning algorithm including its classification. The attributes that define the search space can be poorly understood or inadequate, thereby making it difficult to discover high quality knowledge and understanding. Feature construction (FC) and feature selection (FS) are two pre-processing steps that can be used to improve the feature space quality, by enhancing the classifier performance in terms of accuracy, complexity, speed and interpretability. While FS aims to choose a set of informative features for improving the performance, FC can enhance the classification performance by evolving new features out of the original ones. The evolved features are expected to have more predictive value than the originals that make them up. Over the past few decades, several evolutionary computation (EC) methods have been proposed in the area of FC. This paper gives an overview of the literature on EC for FC. Here, we focus mainly on filter, wrapper and embedded methods, in which the contributions of these different methods are identified. Furthermore, some open challenges and current issues are also discussed in order to identify promising areas for future research.

Original languageEnglish
Title of host publicationStudies in Computational Intelligence
PublisherSpringer Verlag
Pages109-122
Number of pages14
DOIs
Publication statusPublished - 1 Jan 2020

Publication series

NameStudies in Computational Intelligence
Volume830
ISSN (Print)1860-949X

Fingerprint

Evolutionary algorithms
Feature extraction
Learning algorithms
Learning systems
Classifiers
Processing

Keywords

  • Classification
  • Evolutionary computation
  • Feature construction
  • Feature selection

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Swesi, I. M. A. O., & Abu Bakar, A. (2020). Recent developments on evolutionary computation techniques to feature construction. In Studies in Computational Intelligence (pp. 109-122). (Studies in Computational Intelligence; Vol. 830). Springer Verlag. https://doi.org/10.1007/978-3-030-14132-5_9

Recent developments on evolutionary computation techniques to feature construction. / Swesi, Idheba Mohamad Ali O.; Abu Bakar, Azuraliza.

Studies in Computational Intelligence. Springer Verlag, 2020. p. 109-122 (Studies in Computational Intelligence; Vol. 830).

Research output: Chapter in Book/Report/Conference proceedingChapter

Swesi, IMAO & Abu Bakar, A 2020, Recent developments on evolutionary computation techniques to feature construction. in Studies in Computational Intelligence. Studies in Computational Intelligence, vol. 830, Springer Verlag, pp. 109-122. https://doi.org/10.1007/978-3-030-14132-5_9
Swesi IMAO, Abu Bakar A. Recent developments on evolutionary computation techniques to feature construction. In Studies in Computational Intelligence. Springer Verlag. 2020. p. 109-122. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-030-14132-5_9
Swesi, Idheba Mohamad Ali O. ; Abu Bakar, Azuraliza. / Recent developments on evolutionary computation techniques to feature construction. Studies in Computational Intelligence. Springer Verlag, 2020. pp. 109-122 (Studies in Computational Intelligence).
@inbook{be5004d788834fd68ad049f81cbc53fd,
title = "Recent developments on evolutionary computation techniques to feature construction",
abstract = "The quality of the search space is an important factor that influences the performance of any machine learning algorithm including its classification. The attributes that define the search space can be poorly understood or inadequate, thereby making it difficult to discover high quality knowledge and understanding. Feature construction (FC) and feature selection (FS) are two pre-processing steps that can be used to improve the feature space quality, by enhancing the classifier performance in terms of accuracy, complexity, speed and interpretability. While FS aims to choose a set of informative features for improving the performance, FC can enhance the classification performance by evolving new features out of the original ones. The evolved features are expected to have more predictive value than the originals that make them up. Over the past few decades, several evolutionary computation (EC) methods have been proposed in the area of FC. This paper gives an overview of the literature on EC for FC. Here, we focus mainly on filter, wrapper and embedded methods, in which the contributions of these different methods are identified. Furthermore, some open challenges and current issues are also discussed in order to identify promising areas for future research.",
keywords = "Classification, Evolutionary computation, Feature construction, Feature selection",
author = "Swesi, {Idheba Mohamad Ali O.} and {Abu Bakar}, Azuraliza",
year = "2020",
month = "1",
day = "1",
doi = "10.1007/978-3-030-14132-5_9",
language = "English",
series = "Studies in Computational Intelligence",
publisher = "Springer Verlag",
pages = "109--122",
booktitle = "Studies in Computational Intelligence",

}

TY - CHAP

T1 - Recent developments on evolutionary computation techniques to feature construction

AU - Swesi, Idheba Mohamad Ali O.

AU - Abu Bakar, Azuraliza

PY - 2020/1/1

Y1 - 2020/1/1

N2 - The quality of the search space is an important factor that influences the performance of any machine learning algorithm including its classification. The attributes that define the search space can be poorly understood or inadequate, thereby making it difficult to discover high quality knowledge and understanding. Feature construction (FC) and feature selection (FS) are two pre-processing steps that can be used to improve the feature space quality, by enhancing the classifier performance in terms of accuracy, complexity, speed and interpretability. While FS aims to choose a set of informative features for improving the performance, FC can enhance the classification performance by evolving new features out of the original ones. The evolved features are expected to have more predictive value than the originals that make them up. Over the past few decades, several evolutionary computation (EC) methods have been proposed in the area of FC. This paper gives an overview of the literature on EC for FC. Here, we focus mainly on filter, wrapper and embedded methods, in which the contributions of these different methods are identified. Furthermore, some open challenges and current issues are also discussed in order to identify promising areas for future research.

AB - The quality of the search space is an important factor that influences the performance of any machine learning algorithm including its classification. The attributes that define the search space can be poorly understood or inadequate, thereby making it difficult to discover high quality knowledge and understanding. Feature construction (FC) and feature selection (FS) are two pre-processing steps that can be used to improve the feature space quality, by enhancing the classifier performance in terms of accuracy, complexity, speed and interpretability. While FS aims to choose a set of informative features for improving the performance, FC can enhance the classification performance by evolving new features out of the original ones. The evolved features are expected to have more predictive value than the originals that make them up. Over the past few decades, several evolutionary computation (EC) methods have been proposed in the area of FC. This paper gives an overview of the literature on EC for FC. Here, we focus mainly on filter, wrapper and embedded methods, in which the contributions of these different methods are identified. Furthermore, some open challenges and current issues are also discussed in order to identify promising areas for future research.

KW - Classification

KW - Evolutionary computation

KW - Feature construction

KW - Feature selection

UR - http://www.scopus.com/inward/record.url?scp=85063262293&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063262293&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-14132-5_9

DO - 10.1007/978-3-030-14132-5_9

M3 - Chapter

AN - SCOPUS:85063262293

T3 - Studies in Computational Intelligence

SP - 109

EP - 122

BT - Studies in Computational Intelligence

PB - Springer Verlag

ER -