Theoretically optimal distributed anomaly detection

Aleksandar Lazarevic, Nisheeth Srivastava, Ashutosh Tiwari, Josh Isom, Nikunj C. Oza, Jaideep Srivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

A novel general framework for distributed anomaly detection with theoretical performance guarantees is proposed. Our algorithmic approach combines existing anomaly detection procedures with a novel method for computing global statistics using local sufficient statistics. Under a Gaussian assumption, our distributed algorithm is guaranteed to perform as well as its centralized counterpart, a condition we call 'zero information loss'. We further report experimental results on synthetic as well as real-world data to demonstrate the viability of our approach.

Original languageEnglish (US)
Title of host publicationICDM Workshops 2009 - IEEE International Conference on Data Mining
Pages515-520
Number of pages6
DOIs
StatePublished - Dec 1 2009
Event2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009 - Miami, FL, United States
Duration: Dec 6 2009Dec 6 2009

Other

Other2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009
CountryUnited States
CityMiami, FL
Period12/6/0912/6/09

Fingerprint

Statistics
Parallel algorithms

Keywords

  • Anomaly detection
  • Data mining
  • Distributed

Cite this

Lazarevic, A., Srivastava, N., Tiwari, A., Isom, J., Oza, N. C., & Srivastava, J. (2009). Theoretically optimal distributed anomaly detection. In ICDM Workshops 2009 - IEEE International Conference on Data Mining (pp. 515-520). [5360461] https://doi.org/10.1109/ICDMW.2009.40

Theoretically optimal distributed anomaly detection. / Lazarevic, Aleksandar; Srivastava, Nisheeth; Tiwari, Ashutosh; Isom, Josh; Oza, Nikunj C.; Srivastava, Jaideep.

ICDM Workshops 2009 - IEEE International Conference on Data Mining. 2009. p. 515-520 5360461.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lazarevic, A, Srivastava, N, Tiwari, A, Isom, J, Oza, NC & Srivastava, J 2009, Theoretically optimal distributed anomaly detection. in ICDM Workshops 2009 - IEEE International Conference on Data Mining., 5360461, pp. 515-520, 2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009, Miami, FL, United States, 12/6/09. https://doi.org/10.1109/ICDMW.2009.40
Lazarevic A, Srivastava N, Tiwari A, Isom J, Oza NC, Srivastava J. Theoretically optimal distributed anomaly detection. In ICDM Workshops 2009 - IEEE International Conference on Data Mining. 2009. p. 515-520. 5360461 https://doi.org/10.1109/ICDMW.2009.40
Lazarevic, Aleksandar ; Srivastava, Nisheeth ; Tiwari, Ashutosh ; Isom, Josh ; Oza, Nikunj C. ; Srivastava, Jaideep. / Theoretically optimal distributed anomaly detection. ICDM Workshops 2009 - IEEE International Conference on Data Mining. 2009. pp. 515-520
@inproceedings{f4781031427b4159acd167357f645213,
title = "Theoretically optimal distributed anomaly detection",
abstract = "A novel general framework for distributed anomaly detection with theoretical performance guarantees is proposed. Our algorithmic approach combines existing anomaly detection procedures with a novel method for computing global statistics using local sufficient statistics. Under a Gaussian assumption, our distributed algorithm is guaranteed to perform as well as its centralized counterpart, a condition we call 'zero information loss'. We further report experimental results on synthetic as well as real-world data to demonstrate the viability of our approach.",
keywords = "Anomaly detection, Data mining, Distributed",
author = "Aleksandar Lazarevic and Nisheeth Srivastava and Ashutosh Tiwari and Josh Isom and Oza, {Nikunj C.} and Jaideep Srivastava",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/ICDMW.2009.40",
language = "English (US)",
isbn = "9780769539027",
pages = "515--520",
booktitle = "ICDM Workshops 2009 - IEEE International Conference on Data Mining",

}

TY - GEN

T1 - Theoretically optimal distributed anomaly detection

AU - Lazarevic, Aleksandar

AU - Srivastava, Nisheeth

AU - Tiwari, Ashutosh

AU - Isom, Josh

AU - Oza, Nikunj C.

AU - Srivastava, Jaideep

PY - 2009/12/1

Y1 - 2009/12/1

N2 - A novel general framework for distributed anomaly detection with theoretical performance guarantees is proposed. Our algorithmic approach combines existing anomaly detection procedures with a novel method for computing global statistics using local sufficient statistics. Under a Gaussian assumption, our distributed algorithm is guaranteed to perform as well as its centralized counterpart, a condition we call 'zero information loss'. We further report experimental results on synthetic as well as real-world data to demonstrate the viability of our approach.

AB - A novel general framework for distributed anomaly detection with theoretical performance guarantees is proposed. Our algorithmic approach combines existing anomaly detection procedures with a novel method for computing global statistics using local sufficient statistics. Under a Gaussian assumption, our distributed algorithm is guaranteed to perform as well as its centralized counterpart, a condition we call 'zero information loss'. We further report experimental results on synthetic as well as real-world data to demonstrate the viability of our approach.

KW - Anomaly detection

KW - Data mining

KW - Distributed

UR - http://www.scopus.com/inward/record.url?scp=77951166524&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951166524&partnerID=8YFLogxK

U2 - 10.1109/ICDMW.2009.40

DO - 10.1109/ICDMW.2009.40

M3 - Conference contribution

AN - SCOPUS:77951166524

SN - 9780769539027

SP - 515

EP - 520

BT - ICDM Workshops 2009 - IEEE International Conference on Data Mining

ER -