Anomaly detection via over-sampling principal component analysis

Yi Ren Yeh; Zheng Yi Lee; Yuh-Jye Lee

doi:10.1007/978-3-642-00909-9_43

Anomaly detection via over-sampling principal component analysis

Yi Ren Yeh^*, Zheng Yi Lee, Yuh-Jye Lee

^*Corresponding author for this work

Department of Applied Mathematics

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

13 Scopus citations

Abstract

Outlier detection is an important issue in datamining and has been studied in different research areas. It can be used for detecting the small amount of deviated data. In this article, we use "Leave One Out" procedure to check each individual point the "with or without" effect on the variation of principal directions. Based on this idea, an over-sampling principal component analysis outlier detection method is proposed for emphasizing the influence of an abnormal instance (or an outlier). Except for identifying the suspicious outliers, we also design an on-line anomaly detection to detect the new arriving anomaly. In addition, we also study the quick updating of the principal directions for the effective computation and satisfying the on-line detecting demand. Numerical experiments show that our proposed method is effective in computation time and anomaly detection.

Original language	English
Title of host publication	New Advances in Intelligent Decision Technologies
Subtitle of host publication	Results of the First KES International Symposium IDT 2009
Editors	Kazumi Nakamatsu, Gloria Phillips-Wren, Lakhmi Jain, Robert Howlett
Publisher	Springer Verlag
Pages	449-458
Number of pages	10
ISBN (Print)	9783642009082
DOIs	https://doi.org/10.1007/978-3-642-00909-9_43
State	Published - 7 May 2009

Publication series

Name	Studies in Computational Intelligence
Volume	199
ISSN (Print)	1860-949X

Access to Document

10.1007/978-3-642-00909-9_43

Fingerprint Dive into the research topics of 'Anomaly detection via over-sampling principal component analysis'. Together they form a unique fingerprint.

Cite this

Yeh, Y. R., Lee, Z. Y., & Lee, Y-J. (2009). Anomaly detection via over-sampling principal component analysis. In K. Nakamatsu, G. Phillips-Wren, L. Jain, & R. Howlett (Eds.), New Advances in Intelligent Decision Technologies: Results of the First KES International Symposium IDT 2009 (pp. 449-458). (Studies in Computational Intelligence; Vol. 199). Springer Verlag. https://doi.org/10.1007/978-3-642-00909-9_43

Yeh, Yi Ren ; Lee, Zheng Yi ; Lee, Yuh-Jye. / Anomaly detection via over-sampling principal component analysis. New Advances in Intelligent Decision Technologies: Results of the First KES International Symposium IDT 2009. editor / Kazumi Nakamatsu ; Gloria Phillips-Wren ; Lakhmi Jain ; Robert Howlett. Springer Verlag, 2009. pp. 449-458 (Studies in Computational Intelligence).

@inproceedings{a122b3abb39f4519bc271fcf3716c185,

title = "Anomaly detection via over-sampling principal component analysis",

abstract = "Outlier detection is an important issue in datamining and has been studied in different research areas. It can be used for detecting the small amount of deviated data. In this article, we use {"}Leave One Out{"} procedure to check each individual point the {"}with or without{"} effect on the variation of principal directions. Based on this idea, an over-sampling principal component analysis outlier detection method is proposed for emphasizing the influence of an abnormal instance (or an outlier). Except for identifying the suspicious outliers, we also design an on-line anomaly detection to detect the new arriving anomaly. In addition, we also study the quick updating of the principal directions for the effective computation and satisfying the on-line detecting demand. Numerical experiments show that our proposed method is effective in computation time and anomaly detection.",

author = "Yeh, {Yi Ren} and Lee, {Zheng Yi} and Yuh-Jye Lee",

year = "2009",

month = may,

day = "7",

doi = "10.1007/978-3-642-00909-9_43",

language = "English",

isbn = "9783642009082",

series = "Studies in Computational Intelligence",

publisher = "Springer Verlag",

pages = "449--458",

editor = "Kazumi Nakamatsu and Gloria Phillips-Wren and Lakhmi Jain and Robert Howlett",

booktitle = "New Advances in Intelligent Decision Technologies",

address = "Germany",

}

Yeh, YR, Lee, ZY & Lee, Y-J 2009, Anomaly detection via over-sampling principal component analysis. in K Nakamatsu, G Phillips-Wren, L Jain & R Howlett (eds), New Advances in Intelligent Decision Technologies: Results of the First KES International Symposium IDT 2009. Studies in Computational Intelligence, vol. 199, Springer Verlag, pp. 449-458. https://doi.org/10.1007/978-3-642-00909-9_43

Anomaly detection via over-sampling principal component analysis. / Yeh, Yi Ren; Lee, Zheng Yi; Lee, Yuh-Jye.

New Advances in Intelligent Decision Technologies: Results of the First KES International Symposium IDT 2009. ed. / Kazumi Nakamatsu; Gloria Phillips-Wren; Lakhmi Jain; Robert Howlett. Springer Verlag, 2009. p. 449-458 (Studies in Computational Intelligence; Vol. 199).

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

TY - GEN

T1 - Anomaly detection via over-sampling principal component analysis

AU - Yeh, Yi Ren

AU - Lee, Zheng Yi

AU - Lee, Yuh-Jye

PY - 2009/5/7

Y1 - 2009/5/7

N2 - Outlier detection is an important issue in datamining and has been studied in different research areas. It can be used for detecting the small amount of deviated data. In this article, we use "Leave One Out" procedure to check each individual point the "with or without" effect on the variation of principal directions. Based on this idea, an over-sampling principal component analysis outlier detection method is proposed for emphasizing the influence of an abnormal instance (or an outlier). Except for identifying the suspicious outliers, we also design an on-line anomaly detection to detect the new arriving anomaly. In addition, we also study the quick updating of the principal directions for the effective computation and satisfying the on-line detecting demand. Numerical experiments show that our proposed method is effective in computation time and anomaly detection.

AB - Outlier detection is an important issue in datamining and has been studied in different research areas. It can be used for detecting the small amount of deviated data. In this article, we use "Leave One Out" procedure to check each individual point the "with or without" effect on the variation of principal directions. Based on this idea, an over-sampling principal component analysis outlier detection method is proposed for emphasizing the influence of an abnormal instance (or an outlier). Except for identifying the suspicious outliers, we also design an on-line anomaly detection to detect the new arriving anomaly. In addition, we also study the quick updating of the principal directions for the effective computation and satisfying the on-line detecting demand. Numerical experiments show that our proposed method is effective in computation time and anomaly detection.

UR - http://www.scopus.com/inward/record.url?scp=65449122013&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-00909-9_43

DO - 10.1007/978-3-642-00909-9_43

M3 - Conference contribution

AN - SCOPUS:65449122013

SN - 9783642009082

T3 - Studies in Computational Intelligence

SP - 449

EP - 458

BT - New Advances in Intelligent Decision Technologies

A2 - Nakamatsu, Kazumi

A2 - Phillips-Wren, Gloria

A2 - Jain, Lakhmi

A2 - Howlett, Robert

PB - Springer Verlag

ER -

Yeh YR, Lee ZY, Lee Y-J. Anomaly detection via over-sampling principal component analysis. In Nakamatsu K, Phillips-Wren G, Jain L, Howlett R, editors, New Advances in Intelligent Decision Technologies: Results of the First KES International Symposium IDT 2009. Springer Verlag. 2009. p. 449-458. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-642-00909-9_43

Anomaly detection via over-sampling principal component analysis

Abstract

Publication series

Access to Document

Other files and links

Cite this