An error rate comparison of classification methods with continuous explanatory variables

Benjamin J. Nelson, George Runger, Jennie Si

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In the fields of statistics and computer science a wide variety of methodologies exist for solving the traditional classification problem. This study will compare the error rates for various methods under a variety of conditions when the explanatory variables are continuous. The methods under considerations are neural networks, classical discriminant analysis, and two different approaches to decision trees. Training and testing sets are utilized to estimate the error rates of these methods for different numbers of sample sizes, number of explanatory variables, and the number of classes in the dependent variable. These error rates will be used to draw generalized conclusions about the relative efficiencies of the techniques.

Original languageEnglish (US)
Pages (from-to)557-566
Number of pages10
JournalIIE Transactions (Institute of Industrial Engineers)
Volume35
Issue number6
DOIs
StatePublished - Jun 2003

Fingerprint

Discriminant analysis
Decision trees
Computer science
Statistics
Neural networks
Testing
Relative efficiency
Decision tree
Sample size
Methodology

ASJC Scopus subject areas

  • Management Science and Operations Research
  • Industrial and Manufacturing Engineering

Cite this

An error rate comparison of classification methods with continuous explanatory variables. / Nelson, Benjamin J.; Runger, George; Si, Jennie.

In: IIE Transactions (Institute of Industrial Engineers), Vol. 35, No. 6, 06.2003, p. 557-566.

Research output: Contribution to journalArticle

@article{bbdbde94a11a4cf5bd23a214d85910ed,
title = "An error rate comparison of classification methods with continuous explanatory variables",
abstract = "In the fields of statistics and computer science a wide variety of methodologies exist for solving the traditional classification problem. This study will compare the error rates for various methods under a variety of conditions when the explanatory variables are continuous. The methods under considerations are neural networks, classical discriminant analysis, and two different approaches to decision trees. Training and testing sets are utilized to estimate the error rates of these methods for different numbers of sample sizes, number of explanatory variables, and the number of classes in the dependent variable. These error rates will be used to draw generalized conclusions about the relative efficiencies of the techniques.",
author = "Nelson, {Benjamin J.} and George Runger and Jennie Si",
year = "2003",
month = "6",
doi = "10.1080/07408170304421",
language = "English (US)",
volume = "35",
pages = "557--566",
journal = "IISE Transactions",
issn = "2472-5854",
publisher = "Taylor and Francis Ltd.",
number = "6",

}

TY - JOUR

T1 - An error rate comparison of classification methods with continuous explanatory variables

AU - Nelson, Benjamin J.

AU - Runger, George

AU - Si, Jennie

PY - 2003/6

Y1 - 2003/6

N2 - In the fields of statistics and computer science a wide variety of methodologies exist for solving the traditional classification problem. This study will compare the error rates for various methods under a variety of conditions when the explanatory variables are continuous. The methods under considerations are neural networks, classical discriminant analysis, and two different approaches to decision trees. Training and testing sets are utilized to estimate the error rates of these methods for different numbers of sample sizes, number of explanatory variables, and the number of classes in the dependent variable. These error rates will be used to draw generalized conclusions about the relative efficiencies of the techniques.

AB - In the fields of statistics and computer science a wide variety of methodologies exist for solving the traditional classification problem. This study will compare the error rates for various methods under a variety of conditions when the explanatory variables are continuous. The methods under considerations are neural networks, classical discriminant analysis, and two different approaches to decision trees. Training and testing sets are utilized to estimate the error rates of these methods for different numbers of sample sizes, number of explanatory variables, and the number of classes in the dependent variable. These error rates will be used to draw generalized conclusions about the relative efficiencies of the techniques.

UR - http://www.scopus.com/inward/record.url?scp=0038136650&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038136650&partnerID=8YFLogxK

U2 - 10.1080/07408170304421

DO - 10.1080/07408170304421

M3 - Article

AN - SCOPUS:0038136650

VL - 35

SP - 557

EP - 566

JO - IISE Transactions

JF - IISE Transactions

SN - 2472-5854

IS - 6

ER -