Anomalous subgraph detection via Sparse Principal Component Analysis

Navraj Singh, Benjamin A. Miller, Nadya Bliss, Patrick J. Wolfe

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

9 Citations (Scopus)

Abstract

Network datasets have become ubiquitous in many fields of study in recent years. In this paper we investigate a problem with applicability to a wide variety of domains detecting small, anomalous subgraphs in a background graph. We characterize the anomaly in a subgraph via the well-known notion of network modularity, and we show that the optimization problem formulation resulting from our setup is very similar to a recently introduced technique in statistics called Sparse Principal Component Analysis (Sparse PCA), which is an extension of the classical PCA algorithm. The exact version of our problem formulation is a hard combinatorial optimization problem, so we consider a recently introduced semidefinite programming relaxation of the Sparse PCA problem. We show via results on simulated data that the technique is very promising.

Original languageEnglish (US)
Title of host publicationIEEE Workshop on Statistical Signal Processing Proceedings
Pages485-488
Number of pages4
DOIs
StatePublished - 2011
Externally publishedYes
Event2011 IEEE Statistical Signal Processing Workshop, SSP 2011 - Nice, France
Duration: Jun 28 2011Jun 30 2011

Other

Other2011 IEEE Statistical Signal Processing Workshop, SSP 2011
CountryFrance
CityNice
Period6/28/116/30/11

Fingerprint

Principal component analysis
Principal Component Analysis
Anomalous
Subgraph
Combinatorial optimization
Semidefinite Programming Relaxation
Formulation
Statistics
Modularity
Combinatorial Optimization Problem
Anomaly
Optimization Problem
Graph in graph theory
Background

Keywords

  • Anomaly detection
  • community detection
  • graph analysis
  • semidefinite programming
  • sparse principal component analysis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Applied Mathematics
  • Signal Processing
  • Computer Science Applications

Cite this

Singh, N., Miller, B. A., Bliss, N., & Wolfe, P. J. (2011). Anomalous subgraph detection via Sparse Principal Component Analysis. In IEEE Workshop on Statistical Signal Processing Proceedings (pp. 485-488). [5967738] https://doi.org/10.1109/SSP.2011.5967738

Anomalous subgraph detection via Sparse Principal Component Analysis. / Singh, Navraj; Miller, Benjamin A.; Bliss, Nadya; Wolfe, Patrick J.

IEEE Workshop on Statistical Signal Processing Proceedings. 2011. p. 485-488 5967738.

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

Singh, N, Miller, BA, Bliss, N & Wolfe, PJ 2011, Anomalous subgraph detection via Sparse Principal Component Analysis. in IEEE Workshop on Statistical Signal Processing Proceedings., 5967738, pp. 485-488, 2011 IEEE Statistical Signal Processing Workshop, SSP 2011, Nice, France, 6/28/11. https://doi.org/10.1109/SSP.2011.5967738
Singh N, Miller BA, Bliss N, Wolfe PJ. Anomalous subgraph detection via Sparse Principal Component Analysis. In IEEE Workshop on Statistical Signal Processing Proceedings. 2011. p. 485-488. 5967738 https://doi.org/10.1109/SSP.2011.5967738
Singh, Navraj ; Miller, Benjamin A. ; Bliss, Nadya ; Wolfe, Patrick J. / Anomalous subgraph detection via Sparse Principal Component Analysis. IEEE Workshop on Statistical Signal Processing Proceedings. 2011. pp. 485-488
@inproceedings{993074e0da5e44feaaa9138976eda373,
title = "Anomalous subgraph detection via Sparse Principal Component Analysis",
abstract = "Network datasets have become ubiquitous in many fields of study in recent years. In this paper we investigate a problem with applicability to a wide variety of domains detecting small, anomalous subgraphs in a background graph. We characterize the anomaly in a subgraph via the well-known notion of network modularity, and we show that the optimization problem formulation resulting from our setup is very similar to a recently introduced technique in statistics called Sparse Principal Component Analysis (Sparse PCA), which is an extension of the classical PCA algorithm. The exact version of our problem formulation is a hard combinatorial optimization problem, so we consider a recently introduced semidefinite programming relaxation of the Sparse PCA problem. We show via results on simulated data that the technique is very promising.",
keywords = "Anomaly detection, community detection, graph analysis, semidefinite programming, sparse principal component analysis",
author = "Navraj Singh and Miller, {Benjamin A.} and Nadya Bliss and Wolfe, {Patrick J.}",
year = "2011",
doi = "10.1109/SSP.2011.5967738",
language = "English (US)",
isbn = "9781457705700",
pages = "485--488",
booktitle = "IEEE Workshop on Statistical Signal Processing Proceedings",

}

TY - GEN

T1 - Anomalous subgraph detection via Sparse Principal Component Analysis

AU - Singh, Navraj

AU - Miller, Benjamin A.

AU - Bliss, Nadya

AU - Wolfe, Patrick J.

PY - 2011

Y1 - 2011

N2 - Network datasets have become ubiquitous in many fields of study in recent years. In this paper we investigate a problem with applicability to a wide variety of domains detecting small, anomalous subgraphs in a background graph. We characterize the anomaly in a subgraph via the well-known notion of network modularity, and we show that the optimization problem formulation resulting from our setup is very similar to a recently introduced technique in statistics called Sparse Principal Component Analysis (Sparse PCA), which is an extension of the classical PCA algorithm. The exact version of our problem formulation is a hard combinatorial optimization problem, so we consider a recently introduced semidefinite programming relaxation of the Sparse PCA problem. We show via results on simulated data that the technique is very promising.

AB - Network datasets have become ubiquitous in many fields of study in recent years. In this paper we investigate a problem with applicability to a wide variety of domains detecting small, anomalous subgraphs in a background graph. We characterize the anomaly in a subgraph via the well-known notion of network modularity, and we show that the optimization problem formulation resulting from our setup is very similar to a recently introduced technique in statistics called Sparse Principal Component Analysis (Sparse PCA), which is an extension of the classical PCA algorithm. The exact version of our problem formulation is a hard combinatorial optimization problem, so we consider a recently introduced semidefinite programming relaxation of the Sparse PCA problem. We show via results on simulated data that the technique is very promising.

KW - Anomaly detection

KW - community detection

KW - graph analysis

KW - semidefinite programming

KW - sparse principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=80052186985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052186985&partnerID=8YFLogxK

U2 - 10.1109/SSP.2011.5967738

DO - 10.1109/SSP.2011.5967738

M3 - Conference contribution

SN - 9781457705700

SP - 485

EP - 488

BT - IEEE Workshop on Statistical Signal Processing Proceedings

ER -