Self-organizing neural networks for data projection

Mu Chun Su; Hisao Te Chang

doi:10.1007/978-3-540-46652-9_20

Self-organizing neural networks for data projection

Mu Chun Su, Hisao Te Chang

Department of Computer Science and Information Engineering

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

3 Scopus citations

Abstract

In this paper we present a nonlinear projection method for visualizing high-dimensional data as a two-dimensional scatter plot. The method is based on a new model of self-organizing neural networks. An algorithm called “double self-organizing feature map” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its architecture during the learning phase so as to make neurons responding to similar stimulus be clustered together. Then the architecture of the network is graphically displayed to show the underlying structure of the data. Two data sets are used to test the effectiveness of the proposed neural network.

Original language	English
Title of host publication	Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings
Editors	Lucas Chi-Kwong Hui, Dik Lun Lee
Publisher	Springer Verlag
Pages	206-215
Number of pages	10
ISBN (Print)	3540669035, 9783540669036
DOIs	https://doi.org/10.1007/978-3-540-46652-9_20
State	Published - 1999
Event	5th International Computer Science Conference, ICSC 1999 - Hong Kong, China Duration: 13 Dec 1999 → 15 Dec 1999

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	1749
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Computer Science Conference, ICSC 1999
Country/Territory	China
City	Hong Kong
Period	13/12/99 → 15/12/99

Access to Document

10.1007/978-3-540-46652-9_20

Cite this

Su, M. C., & Chang, H. T. (1999). Self-organizing neural networks for data projection. In L. C.-K. Hui, & D. L. Lee (Eds.), Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings (pp. 206-215). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1749). Springer Verlag. https://doi.org/10.1007/978-3-540-46652-9_20

Su, Mu Chun ; Chang, Hisao Te. / Self-organizing neural networks for data projection. Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings. editor / Lucas Chi-Kwong Hui ; Dik Lun Lee. Springer Verlag, 1999. pp. 206-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{9420328bbe514701b707bc851cceaedb,

title = "Self-organizing neural networks for data projection",

abstract = "In this paper we present a nonlinear projection method for visualizing high-dimensional data as a two-dimensional scatter plot. The method is based on a new model of self-organizing neural networks. An algorithm called “double self-organizing feature map” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its architecture during the learning phase so as to make neurons responding to similar stimulus be clustered together. Then the architecture of the network is graphically displayed to show the underlying structure of the data. Two data sets are used to test the effectiveness of the proposed neural network.",

author = "Su, {Mu Chun} and Chang, {Hisao Te}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 1999.; 5th International Computer Science Conference, ICSC 1999 ; Conference date: 13-12-1999 Through 15-12-1999",

year = "1999",

doi = "10.1007/978-3-540-46652-9_20",

language = "???core.languages.en_GB???",

isbn = "3540669035",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "206--215",

editor = "Hui, {Lucas Chi-Kwong} and Lee, {Dik Lun}",

booktitle = "Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings",

}

Su, MC & Chang, HT 1999, Self-organizing neural networks for data projection. in LC-K Hui & DL Lee (eds), Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1749, Springer Verlag, pp. 206-215, 5th International Computer Science Conference, ICSC 1999, Hong Kong, China, 13/12/99. https://doi.org/10.1007/978-3-540-46652-9_20

Self-organizing neural networks for data projection. / Su, Mu Chun; Chang, Hisao Te.
Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings. ed. / Lucas Chi-Kwong Hui; Dik Lun Lee. Springer Verlag, 1999. p. 206-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1749).

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

TY - GEN

T1 - Self-organizing neural networks for data projection

AU - Su, Mu Chun

AU - Chang, Hisao Te

N1 - Publisher Copyright: © Springer-Verlag Berlin Heidelberg 1999.

PY - 1999

Y1 - 1999

N2 - In this paper we present a nonlinear projection method for visualizing high-dimensional data as a two-dimensional scatter plot. The method is based on a new model of self-organizing neural networks. An algorithm called “double self-organizing feature map” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its architecture during the learning phase so as to make neurons responding to similar stimulus be clustered together. Then the architecture of the network is graphically displayed to show the underlying structure of the data. Two data sets are used to test the effectiveness of the proposed neural network.

AB - In this paper we present a nonlinear projection method for visualizing high-dimensional data as a two-dimensional scatter plot. The method is based on a new model of self-organizing neural networks. An algorithm called “double self-organizing feature map” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its architecture during the learning phase so as to make neurons responding to similar stimulus be clustered together. Then the architecture of the network is graphically displayed to show the underlying structure of the data. Two data sets are used to test the effectiveness of the proposed neural network.

UR - http://www.scopus.com/inward/record.url?scp=84947804367&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-46652-9_20

DO - 10.1007/978-3-540-46652-9_20

M3 - 會議論文篇章

AN - SCOPUS:84947804367

SN - 3540669035

SN - 9783540669036

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 206

EP - 215

BT - Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings

A2 - Hui, Lucas Chi-Kwong

A2 - Lee, Dik Lun

PB - Springer Verlag

T2 - 5th International Computer Science Conference, ICSC 1999

Y2 - 13 December 1999 through 15 December 1999

ER -

Su MC, Chang HT. Self-organizing neural networks for data projection. In Hui LCK, Lee DL, editors, Internet Applications - 5th International Computer Science Conference ICSC 1999, Proceedings. Springer Verlag. 1999. p. 206-215. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-540-46652-9_20

Self-organizing neural networks for data projection

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this