Evolutionary-modified fuzzy nearest-neighbor rule for pattern classification

Peyman Hosseinzadeh Kassani, Andrew Beng Jin Teoh, Euntai Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper presents an improved version of the well-established k nearest neighbor (k-NN) and fuzzy NN (FNN), termed the multi-objective genetic-algorithm-modified FNN (MOGA-MFNN). The MFNN design problem is converted into a multi-modal objective maximization problem constrained by four objective functions. Thereafter, the associated parameter set of the MFNN and the feature attributes can be determined optimally and automatically via the non-dominated sorting genetic algorithm II. We introduce two new objective functions termed the Margin-I and Margin-II, which are used to improve the generalization capability of the MFNN for the unknown data, along with two existing performance functions: the geometric mean and the area under the receiver-operated characteristic curve for the training accuracy. Moreover, we proposed a novel data-dependent weight-assignment technique for local class membership functions of the MFNN. The technique enables the MFNN to determine its local neighbors adaptively through the MOGA algorithm. To expedite the classification, the MOGA-MFNN is implemented on a graphical processing unit (GPU), which significantly increases the computation speed. Furthermore, the local class-membership function of the MFNN can be computed in advance, rather than delaying it to the classification stage. This again can improve the classification speed. The MOGA-MFNN is evaluated on 20 datasets obtained from the repository of the University of California, Irvine (UCI). The experiments with rigorous statistical significance tests demonstrate that the proposed method performs competitively with the existing methods.

Original languageEnglish
Pages (from-to)258-269
Number of pages12
JournalExpert Systems with Applications
Volume88
DOIs
Publication statusPublished - 2017 Dec 1

Fingerprint

Pattern recognition
Genetic algorithms
Membership functions
Statistical tests
Sorting
Processing
Experiments

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

@article{2b479a6ee84f464a9470bbbf06b7dc9f,
title = "Evolutionary-modified fuzzy nearest-neighbor rule for pattern classification",
abstract = "This paper presents an improved version of the well-established k nearest neighbor (k-NN) and fuzzy NN (FNN), termed the multi-objective genetic-algorithm-modified FNN (MOGA-MFNN). The MFNN design problem is converted into a multi-modal objective maximization problem constrained by four objective functions. Thereafter, the associated parameter set of the MFNN and the feature attributes can be determined optimally and automatically via the non-dominated sorting genetic algorithm II. We introduce two new objective functions termed the Margin-I and Margin-II, which are used to improve the generalization capability of the MFNN for the unknown data, along with two existing performance functions: the geometric mean and the area under the receiver-operated characteristic curve for the training accuracy. Moreover, we proposed a novel data-dependent weight-assignment technique for local class membership functions of the MFNN. The technique enables the MFNN to determine its local neighbors adaptively through the MOGA algorithm. To expedite the classification, the MOGA-MFNN is implemented on a graphical processing unit (GPU), which significantly increases the computation speed. Furthermore, the local class-membership function of the MFNN can be computed in advance, rather than delaying it to the classification stage. This again can improve the classification speed. The MOGA-MFNN is evaluated on 20 datasets obtained from the repository of the University of California, Irvine (UCI). The experiments with rigorous statistical significance tests demonstrate that the proposed method performs competitively with the existing methods.",
author = "Kassani, {Peyman Hosseinzadeh} and Teoh, {Andrew Beng Jin} and Euntai Kim",
year = "2017",
month = "12",
day = "1",
doi = "10.1016/j.eswa.2017.07.013",
language = "English",
volume = "88",
pages = "258--269",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

Evolutionary-modified fuzzy nearest-neighbor rule for pattern classification. / Kassani, Peyman Hosseinzadeh; Teoh, Andrew Beng Jin; Kim, Euntai.

In: Expert Systems with Applications, Vol. 88, 01.12.2017, p. 258-269.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Evolutionary-modified fuzzy nearest-neighbor rule for pattern classification

AU - Kassani, Peyman Hosseinzadeh

AU - Teoh, Andrew Beng Jin

AU - Kim, Euntai

PY - 2017/12/1

Y1 - 2017/12/1

N2 - This paper presents an improved version of the well-established k nearest neighbor (k-NN) and fuzzy NN (FNN), termed the multi-objective genetic-algorithm-modified FNN (MOGA-MFNN). The MFNN design problem is converted into a multi-modal objective maximization problem constrained by four objective functions. Thereafter, the associated parameter set of the MFNN and the feature attributes can be determined optimally and automatically via the non-dominated sorting genetic algorithm II. We introduce two new objective functions termed the Margin-I and Margin-II, which are used to improve the generalization capability of the MFNN for the unknown data, along with two existing performance functions: the geometric mean and the area under the receiver-operated characteristic curve for the training accuracy. Moreover, we proposed a novel data-dependent weight-assignment technique for local class membership functions of the MFNN. The technique enables the MFNN to determine its local neighbors adaptively through the MOGA algorithm. To expedite the classification, the MOGA-MFNN is implemented on a graphical processing unit (GPU), which significantly increases the computation speed. Furthermore, the local class-membership function of the MFNN can be computed in advance, rather than delaying it to the classification stage. This again can improve the classification speed. The MOGA-MFNN is evaluated on 20 datasets obtained from the repository of the University of California, Irvine (UCI). The experiments with rigorous statistical significance tests demonstrate that the proposed method performs competitively with the existing methods.

AB - This paper presents an improved version of the well-established k nearest neighbor (k-NN) and fuzzy NN (FNN), termed the multi-objective genetic-algorithm-modified FNN (MOGA-MFNN). The MFNN design problem is converted into a multi-modal objective maximization problem constrained by four objective functions. Thereafter, the associated parameter set of the MFNN and the feature attributes can be determined optimally and automatically via the non-dominated sorting genetic algorithm II. We introduce two new objective functions termed the Margin-I and Margin-II, which are used to improve the generalization capability of the MFNN for the unknown data, along with two existing performance functions: the geometric mean and the area under the receiver-operated characteristic curve for the training accuracy. Moreover, we proposed a novel data-dependent weight-assignment technique for local class membership functions of the MFNN. The technique enables the MFNN to determine its local neighbors adaptively through the MOGA algorithm. To expedite the classification, the MOGA-MFNN is implemented on a graphical processing unit (GPU), which significantly increases the computation speed. Furthermore, the local class-membership function of the MFNN can be computed in advance, rather than delaying it to the classification stage. This again can improve the classification speed. The MOGA-MFNN is evaluated on 20 datasets obtained from the repository of the University of California, Irvine (UCI). The experiments with rigorous statistical significance tests demonstrate that the proposed method performs competitively with the existing methods.

UR - http://www.scopus.com/inward/record.url?scp=85023613963&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023613963&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2017.07.013

DO - 10.1016/j.eswa.2017.07.013

M3 - Article

AN - SCOPUS:85023613963

VL - 88

SP - 258

EP - 269

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -