Accuracy estimation for supervised learning algorithms

Charles W. Glover; Ed M. Oblow; Nageswara S. Rao

Accuracy estimation for supervised learning algorithms

Charles W. Glover
, Ed M. Oblow
, Nageswara S. Rao

Autonomous and Complex Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper illustrates and discusses the relative merits of three methods - k-fold Cross Validation, Error Bounds, and Incremental Halting Test - to estimate the accuracy of a supervised learning algorithm. For each of the three methods we point out the problem they address, some of the important assumptions that they are based on, and illustrate them through an example. Finally, we discuss the relative advantages and disadvantages of each method.

Original language	English
Title of host publication	Proceedings of SPIE - The International Society for Optical Engineering
Editors	Steven K. Rogers
Publisher	Society of Photo-Optical Instrumentation Engineers
Pages	794-802
Number of pages	9
ISBN (Print)	0819424927
State	Published - 1997
Event	Applications and Science of Artificial Neural Networks III - Orlando, FL, USA Duration: Apr 21 1997 → Apr 24 1997

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	3077
ISSN (Print)	0277-786X

Conference

Conference	Applications and Science of Artificial Neural Networks III
City	Orlando, FL, USA
Period	04/21/97 → 04/24/97

Cite this

@inproceedings{046e3e955cf24e78ba3fca7afa0de66a,

title = "Accuracy estimation for supervised learning algorithms",

abstract = "This paper illustrates and discusses the relative merits of three methods - k-fold Cross Validation, Error Bounds, and Incremental Halting Test - to estimate the accuracy of a supervised learning algorithm. For each of the three methods we point out the problem they address, some of the important assumptions that they are based on, and illustrate them through an example. Finally, we discuss the relative advantages and disadvantages of each method.",

author = "Glover, \{Charles W.\} and Oblow, \{Ed M.\} and Rao, \{Nageswara S.\}",

year = "1997",

language = "English",

isbn = "0819424927",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "Society of Photo-Optical Instrumentation Engineers",

pages = "794--802",

editor = "Rogers, \{Steven K.\}",

booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

note = "Applications and Science of Artificial Neural Networks III ; Conference date: 21-04-1997 Through 24-04-1997",

}

Glover, CW, Oblow, EM & Rao, NS 1997, Accuracy estimation for supervised learning algorithms. in SK Rogers (ed.), Proceedings of SPIE - The International Society for Optical Engineering. Proceedings of SPIE - The International Society for Optical Engineering, vol. 3077, Society of Photo-Optical Instrumentation Engineers, pp. 794-802, Applications and Science of Artificial Neural Networks III, Orlando, FL, USA, 04/21/97.

Accuracy estimation for supervised learning algorithms. / Glover, Charles W.; Oblow, Ed M.; Rao, Nageswara S.
Proceedings of SPIE - The International Society for Optical Engineering. ed. / Steven K. Rogers. Society of Photo-Optical Instrumentation Engineers, 1997. p. 794-802 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 3077).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Accuracy estimation for supervised learning algorithms

AU - Glover, Charles W.

AU - Oblow, Ed M.

AU - Rao, Nageswara S.

PY - 1997

Y1 - 1997

N2 - This paper illustrates and discusses the relative merits of three methods - k-fold Cross Validation, Error Bounds, and Incremental Halting Test - to estimate the accuracy of a supervised learning algorithm. For each of the three methods we point out the problem they address, some of the important assumptions that they are based on, and illustrate them through an example. Finally, we discuss the relative advantages and disadvantages of each method.

AB - This paper illustrates and discusses the relative merits of three methods - k-fold Cross Validation, Error Bounds, and Incremental Halting Test - to estimate the accuracy of a supervised learning algorithm. For each of the three methods we point out the problem they address, some of the important assumptions that they are based on, and illustrate them through an example. Finally, we discuss the relative advantages and disadvantages of each method.

UR - https://www.scopus.com/pages/publications/0031376891

M3 - Conference contribution

AN - SCOPUS:0031376891

SN - 0819424927

T3 - Proceedings of SPIE - The International Society for Optical Engineering

SP - 794

EP - 802

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Rogers, Steven K.

PB - Society of Photo-Optical Instrumentation Engineers

T2 - Applications and Science of Artificial Neural Networks III

Y2 - 21 April 1997 through 24 April 1997

ER -

Accuracy estimation for supervised learning algorithms

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this