Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi

Mete Ozay; Fatoş T.Yarman Vural

doi:10.1109/SIU.2009.5136569

Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi

Translated title of the contribution: Linear separability analysis for stacked generalization architecture

Mete Ozay, Fatoş T.Yarman Vural

INF - System Information Sciences

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

Abstract

Stacked Generalization algorithm aims to increase the individual classification performances of the classifiers by combining the information obtained from various classifiers in a multilayer architecture by either linear or nonlinear techniques. Performance of the algorithm varies depending on the application domains and the space analyses that affect the classification performances could not be applied successfully. In the present work, linear and nonlinear transformations are investigated within and between each layer, and the linear separability property of the architecture is examined. In the conclusion of the analyses, it is observed that the data space can be separated linearly.

Translated title of the contribution	Linear separability analysis for stacked generalization architecture
Original language	Turkish
Title of host publication	2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009
Pages	1009-1012
Number of pages	4
DOIs	https://doi.org/10.1109/SIU.2009.5136569
Publication status	Published - 2009 Oct 29
Event	2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009 - Antalya, Turkey Duration: 2009 Apr 9 → 2009 Apr 11

Publication series

Name	2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009

Conference

Conference
Country	Turkey
City	Antalya
Period	09/4/9 → 09/4/11

ASJC Scopus subject areas

Computer Networks and Communications
Signal Processing
Communication

Access to Document

10.1109/SIU.2009.5136569

Fingerprint Dive into the research topics of 'Linear separability analysis for stacked generalization architecture'. Together they form a unique fingerprint.

Cite this

Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi. / Ozay, Mete; Vural, Fatoş T.Yarman.

2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009. 2009. p. 1009-1012 5136569 (2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009).

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

Ozay, M & Vural, FTY 2009, Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi. in 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009., 5136569, 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009, pp. 1009-1012, 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009, Antalya, Turkey, 09/4/9. https://doi.org/10.1109/SIU.2009.5136569

@inproceedings{7a1fdc57b25c4a39a23c4af115707621,

title = "Yiǧilmi{\c s} genelleme algoritmasi i{\c c}in doǧrusal ayrilabilirlik analizi",

abstract = "Stacked Generalization algorithm aims to increase the individual classification performances of the classifiers by combining the information obtained from various classifiers in a multilayer architecture by either linear or nonlinear techniques. Performance of the algorithm varies depending on the application domains and the space analyses that affect the classification performances could not be applied successfully. In the present work, linear and nonlinear transformations are investigated within and between each layer, and the linear separability property of the architecture is examined. In the conclusion of the analyses, it is observed that the data space can be separated linearly.",

author = "Mete Ozay and Vural, {Fato{\c s} T.Yarman}",

year = "2009",

month = oct,

day = "29",

doi = "10.1109/SIU.2009.5136569",

language = "Turkish",

isbn = "9781424444366",

series = "2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009",

pages = "1009--1012",

booktitle = "2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009",

note = "2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009 ; Conference date: 09-04-2009 Through 11-04-2009",

}

TY - GEN

T1 - Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi

AU - Ozay, Mete

AU - Vural, Fatoş T.Yarman

PY - 2009/10/29

Y1 - 2009/10/29

N2 - Stacked Generalization algorithm aims to increase the individual classification performances of the classifiers by combining the information obtained from various classifiers in a multilayer architecture by either linear or nonlinear techniques. Performance of the algorithm varies depending on the application domains and the space analyses that affect the classification performances could not be applied successfully. In the present work, linear and nonlinear transformations are investigated within and between each layer, and the linear separability property of the architecture is examined. In the conclusion of the analyses, it is observed that the data space can be separated linearly.

AB - Stacked Generalization algorithm aims to increase the individual classification performances of the classifiers by combining the information obtained from various classifiers in a multilayer architecture by either linear or nonlinear techniques. Performance of the algorithm varies depending on the application domains and the space analyses that affect the classification performances could not be applied successfully. In the present work, linear and nonlinear transformations are investigated within and between each layer, and the linear separability property of the architecture is examined. In the conclusion of the analyses, it is observed that the data space can be separated linearly.

UR - http://www.scopus.com/inward/record.url?scp=70350336534&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350336534&partnerID=8YFLogxK

U2 - 10.1109/SIU.2009.5136569

DO - 10.1109/SIU.2009.5136569

M3 - Conference contribution

AN - SCOPUS:70350336534

SN - 9781424444366

T3 - 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009

SP - 1009

EP - 1012

BT - 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009

T2 - 2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009

Y2 - 9 April 2009 through 11 April 2009

ER -

Yiǧilmiş genelleme algoritmasi için doǧrusal ayrilabilirlik analizi

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this