|Table of Contents|

Implementation of Adapting Aglorithm of Anomalies Detection Based on PCA(PDF)

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2008年04期
Page:
13-16
Research Field:
Publishing date:

Info

Title:
Implementation of Adapting Aglorithm of Anomalies Detection Based on PCA
Author(s):
Fu Qiang; Gan Liang; Li Aiping; Wu Quanyuan
School of Computer Science,National University of Defense Technology,Changsha 410073,China
Keywords:
network anom a ly;  PCA;  adap ting ag lo rithm
PACS:
TP393.08
DOI:
-
Abstract:
A im ing a t the ne tw ork anoma ly detection of distributed ne tw orks, the paper suggests am ultid im ensiona l adapting algor ithm o f anom a lies detection. The a lgor ithm, based on PCA, can change itse lf automa tica lly fo r new anom a ly. H av ing processed netwo rk flow data by PCA algor ithm s, w e can g et the d im ensions w hich contribute most and g ive the re lation be tw een anom alies and d im ensions, and m ake se l-f adaption of features. The exper im enta l results show tha t the a lgo rithm reduces the co st o f netwo rk anom a ly detection, and improves the alert precision of netw ork anom aly de tection

References:

[ 1] H uang Ling, Nguye Xuan long, M inos Garofa lak is, et a.l Comm un ication-e fficient on line de tection o f netw ork-w ide anom aliesk
[ C] / / Proceedings o f the 26th IEEE In ternational Conference on Com puter Comm un ica tions. Ancho rage, AK: IEEE Com puter Soc iety Press, 2007: 134-142.
[ 2] Li Xiao le,i H an Jiaw e.i M in ing approx im ate Top-K subspace anom a lies in m ultidim ensiona l tim eser ies data[ C] / / Proceed ing  o f the 33rd Internationa l Con ference on Very Large Data Bases. V ienna, Austria: VLDB Endowm ent, 2007: 447-458.
[ 3] L iX in, B ian Fang, C rove llaM ark, et a.l Detection and identification o f netwo rk anom a lies using sketch subspaces[ C] / / Proceedings of the 6th ACM S IGCOMM Conference on Interne tM easurem ent. N ew Yrok, USA: ACM, 2006: 147-152.

Memo

Memo:
-
Last Update: 2013-04-24