Algorithm learning based neural network integrating feature selection and classification

Hyunsoo Yoon, Cheong Sool Park, Jun Seok Kim, Jun-Geol Baek

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

Feature selection and classification techniques have been studied independently without considering the interaction between both procedures, which leads to a degraded performance. In this paper, we present a new neural network approach, which is called an algorithm learning based neural network (ALBNN), to improve classification accuracy by integrating feature selection and classification procedures. In general, a knowledge-based artificial neural network operates on prior knowledge from domain experience, which provides it with better starting points for the target function and leads to better classification accuracy. However, prior knowledge is usually difficult to identify. Instead of using unknown background resources, the proposed method utilizes prior knowledge that is mathematically calculated from the properties of other learning algorithms such as PCA, LARS, C4.5, and SVM. We employ the extreme learning machine in this study to help obtain better initial points faster and avoid irrelevant time-consuming work, such as determining architecture and manual tuning. ALBNN correctly approximates a target hypothesis by both considering the interaction between two procedures and minimizing individual procedure errors. The approach produces new relevant features and improves the classification accuracy. Experimental results exhibit improved performance in various classification problems. ALBNN can be applied to various fields requiring high classification accuracy.

Original languageEnglish
Pages (from-to)231-241
Number of pages11
JournalExpert Systems with Applications
Volume40
Issue number1
DOIs
Publication statusPublished - 2013 Jan 1

Fingerprint

Learning algorithms
Feature extraction
Neural networks
Learning systems
Tuning

Keywords

  • Algorithm learning based neural network (ALBNN)
  • Classification
  • Extreme learning machine
  • Feature selection
  • Neural network

ASJC Scopus subject areas

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

Cite this

Algorithm learning based neural network integrating feature selection and classification. / Yoon, Hyunsoo; Park, Cheong Sool; Kim, Jun Seok; Baek, Jun-Geol.

In: Expert Systems with Applications, Vol. 40, No. 1, 01.01.2013, p. 231-241.

Research output: Contribution to journalArticle

Yoon, Hyunsoo ; Park, Cheong Sool ; Kim, Jun Seok ; Baek, Jun-Geol. / Algorithm learning based neural network integrating feature selection and classification. In: Expert Systems with Applications. 2013 ; Vol. 40, No. 1. pp. 231-241.
@article{704015c031a44947912cefc853640a6c,
title = "Algorithm learning based neural network integrating feature selection and classification",
abstract = "Feature selection and classification techniques have been studied independently without considering the interaction between both procedures, which leads to a degraded performance. In this paper, we present a new neural network approach, which is called an algorithm learning based neural network (ALBNN), to improve classification accuracy by integrating feature selection and classification procedures. In general, a knowledge-based artificial neural network operates on prior knowledge from domain experience, which provides it with better starting points for the target function and leads to better classification accuracy. However, prior knowledge is usually difficult to identify. Instead of using unknown background resources, the proposed method utilizes prior knowledge that is mathematically calculated from the properties of other learning algorithms such as PCA, LARS, C4.5, and SVM. We employ the extreme learning machine in this study to help obtain better initial points faster and avoid irrelevant time-consuming work, such as determining architecture and manual tuning. ALBNN correctly approximates a target hypothesis by both considering the interaction between two procedures and minimizing individual procedure errors. The approach produces new relevant features and improves the classification accuracy. Experimental results exhibit improved performance in various classification problems. ALBNN can be applied to various fields requiring high classification accuracy.",
keywords = "Algorithm learning based neural network (ALBNN), Classification, Extreme learning machine, Feature selection, Neural network",
author = "Hyunsoo Yoon and Park, {Cheong Sool} and Kim, {Jun Seok} and Jun-Geol Baek",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.eswa.2012.07.018",
language = "English",
volume = "40",
pages = "231--241",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "1",

}

TY - JOUR

T1 - Algorithm learning based neural network integrating feature selection and classification

AU - Yoon, Hyunsoo

AU - Park, Cheong Sool

AU - Kim, Jun Seok

AU - Baek, Jun-Geol

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Feature selection and classification techniques have been studied independently without considering the interaction between both procedures, which leads to a degraded performance. In this paper, we present a new neural network approach, which is called an algorithm learning based neural network (ALBNN), to improve classification accuracy by integrating feature selection and classification procedures. In general, a knowledge-based artificial neural network operates on prior knowledge from domain experience, which provides it with better starting points for the target function and leads to better classification accuracy. However, prior knowledge is usually difficult to identify. Instead of using unknown background resources, the proposed method utilizes prior knowledge that is mathematically calculated from the properties of other learning algorithms such as PCA, LARS, C4.5, and SVM. We employ the extreme learning machine in this study to help obtain better initial points faster and avoid irrelevant time-consuming work, such as determining architecture and manual tuning. ALBNN correctly approximates a target hypothesis by both considering the interaction between two procedures and minimizing individual procedure errors. The approach produces new relevant features and improves the classification accuracy. Experimental results exhibit improved performance in various classification problems. ALBNN can be applied to various fields requiring high classification accuracy.

AB - Feature selection and classification techniques have been studied independently without considering the interaction between both procedures, which leads to a degraded performance. In this paper, we present a new neural network approach, which is called an algorithm learning based neural network (ALBNN), to improve classification accuracy by integrating feature selection and classification procedures. In general, a knowledge-based artificial neural network operates on prior knowledge from domain experience, which provides it with better starting points for the target function and leads to better classification accuracy. However, prior knowledge is usually difficult to identify. Instead of using unknown background resources, the proposed method utilizes prior knowledge that is mathematically calculated from the properties of other learning algorithms such as PCA, LARS, C4.5, and SVM. We employ the extreme learning machine in this study to help obtain better initial points faster and avoid irrelevant time-consuming work, such as determining architecture and manual tuning. ALBNN correctly approximates a target hypothesis by both considering the interaction between two procedures and minimizing individual procedure errors. The approach produces new relevant features and improves the classification accuracy. Experimental results exhibit improved performance in various classification problems. ALBNN can be applied to various fields requiring high classification accuracy.

KW - Algorithm learning based neural network (ALBNN)

KW - Classification

KW - Extreme learning machine

KW - Feature selection

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=84866123983&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866123983&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2012.07.018

DO - 10.1016/j.eswa.2012.07.018

M3 - Article

AN - SCOPUS:84866123983

VL - 40

SP - 231

EP - 241

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 1

ER -