Combining SOM and local minimum enclosing spheres for novelty detection

Hong-Jie XING; Ming-Hu HA; Xi-Zhao WANG

doi:10.1109/CCDC.2009.5191676

Combining SOM and local minimum enclosing spheres for novelty detection

Hong-Jie XING^*, Ming-Hu HA, Xi-Zhao WANG

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

Abstract

In this paper, a novelty detection method based on self-organizing map (SOM) and local minimum enclosing spheres is proposed. There are two phases in the proposed approach. In the first phase, the whole training set are split into disjointed Voronoi regions by SOM. In the second phase, several local minimum enclosing spheres are constructed upon these Voronoi regions. Compared with its related works, the proposed method demonstrates better performances on one synthetic data set and two benchmark data sets.

Original language	English
Title of host publication	Proceedings : 2009 Chinese Control and Decision Conference, CCDC 2009
Publisher	IEEE
Pages	3771-3776
Number of pages	6
ISBN (Print)	9781424427222
DOIs	https://doi.org/10.1109/CCDC.2009.5191676
Publication status	Published - 2009
Externally published	Yes
Event	2009 Chinese Control and Decision Conference, CCDC 2009 - Guilin, China Duration: 17 Jun 2009 → 19 Jun 2009

Publication series

Name	Chinese Control and Decision Conference, CCDC
Publisher	IEEE
ISSN (Print)	1948-9439
ISSN (Electronic)	1948-9447

Conference

Conference	2009 Chinese Control and Decision Conference, CCDC 2009
Country/Territory	China
City	Guilin
Period	17/06/09 → 19/06/09

Keywords

Local minimum enclosing spheres
Novelty detection
Self-organizing map

Access to Document

10.1109/CCDC.2009.5191676

Cite this

@inproceedings{ea01280d6445465c97ace022c89a8649,

title = "Combining SOM and local minimum enclosing spheres for novelty detection",

abstract = "In this paper, a novelty detection method based on self-organizing map (SOM) and local minimum enclosing spheres is proposed. There are two phases in the proposed approach. In the first phase, the whole training set are split into disjointed Voronoi regions by SOM. In the second phase, several local minimum enclosing spheres are constructed upon these Voronoi regions. Compared with its related works, the proposed method demonstrates better performances on one synthetic data set and two benchmark data sets.",

keywords = "Local minimum enclosing spheres, Novelty detection, Self-organizing map",

author = "Hong-Jie XING and Ming-Hu HA and Xi-Zhao WANG",

year = "2009",

doi = "10.1109/CCDC.2009.5191676",

language = "English",

isbn = "9781424427222",

series = "Chinese Control and Decision Conference, CCDC",

publisher = "IEEE",

pages = "3771--3776",

booktitle = "Proceedings : 2009 Chinese Control and Decision Conference, CCDC 2009",

note = "2009 Chinese Control and Decision Conference, CCDC 2009 ; Conference date: 17-06-2009 Through 19-06-2009",

}

XING, H-J, HA, M-H & WANG, X-Z 2009, Combining SOM and local minimum enclosing spheres for novelty detection. in Proceedings : 2009 Chinese Control and Decision Conference, CCDC 2009., 5191676, Chinese Control and Decision Conference, CCDC, IEEE, pp. 3771-3776, 2009 Chinese Control and Decision Conference, CCDC 2009, Guilin, China, 17/06/09. https://doi.org/10.1109/CCDC.2009.5191676

Combining SOM and local minimum enclosing spheres for novelty detection. / XING, Hong-Jie; HA, Ming-Hu; WANG, Xi-Zhao.
Proceedings : 2009 Chinese Control and Decision Conference, CCDC 2009. IEEE, 2009. p. 3771-3776 5191676 (Chinese Control and Decision Conference, CCDC).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

TY - GEN

T1 - Combining SOM and local minimum enclosing spheres for novelty detection

AU - XING, Hong-Jie

AU - HA, Ming-Hu

AU - WANG, Xi-Zhao

PY - 2009

Y1 - 2009

N2 - In this paper, a novelty detection method based on self-organizing map (SOM) and local minimum enclosing spheres is proposed. There are two phases in the proposed approach. In the first phase, the whole training set are split into disjointed Voronoi regions by SOM. In the second phase, several local minimum enclosing spheres are constructed upon these Voronoi regions. Compared with its related works, the proposed method demonstrates better performances on one synthetic data set and two benchmark data sets.

AB - In this paper, a novelty detection method based on self-organizing map (SOM) and local minimum enclosing spheres is proposed. There are two phases in the proposed approach. In the first phase, the whole training set are split into disjointed Voronoi regions by SOM. In the second phase, several local minimum enclosing spheres are constructed upon these Voronoi regions. Compared with its related works, the proposed method demonstrates better performances on one synthetic data set and two benchmark data sets.

KW - Local minimum enclosing spheres

KW - Novelty detection

KW - Self-organizing map

UR - http://www.scopus.com/inward/record.url?scp=70449347601&partnerID=8YFLogxK

U2 - 10.1109/CCDC.2009.5191676

DO - 10.1109/CCDC.2009.5191676

M3 - Conference paper (refereed)

AN - SCOPUS:70449347601

SN - 9781424427222

T3 - Chinese Control and Decision Conference, CCDC

SP - 3771

EP - 3776

BT - Proceedings : 2009 Chinese Control and Decision Conference, CCDC 2009

PB - IEEE

T2 - 2009 Chinese Control and Decision Conference, CCDC 2009

Y2 - 17 June 2009 through 19 June 2009

ER -

Combining SOM and local minimum enclosing spheres for novelty detection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this