Toward signal processing theory for graphs and non-Euclidean data

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

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

36 Citations (Scopus)

Abstract

Graphs are canonical examples of high-dimensional non-Euclidean data sets, and are emerging as a common data structure in many fields. While there are many algorithms to analyze such data, a signal processing theory for evaluating these techniques akin to detection and estimation in the classical Euclidean setting remains to be developed. In this paper we show the conceptual advantages gained by formulating graph analysis problems in a signal processing framework by way of a practical example: detection of a subgraph embedded in a background graph. We describe an approach based on detection theory and provide empirical results indicating that the test statistic proposed has reasonable power to detect dense subgraphs in large random graphs.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages5414-5417
Number of pages4
DOIs
StatePublished - 2010
Externally publishedYes
Event2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States
Duration: Mar 14 2010Mar 19 2010

Other

Other2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
CountryUnited States
CityDallas, TX
Period3/14/103/19/10

Fingerprint

Signal processing
Data structures
Statistics

Keywords

  • Chi-squared test
  • Community detection
  • Graph algorithms
  • High-dimensional data
  • Signal detection theory

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Miller, B. A., Bliss, N., & Wolfe, P. J. (2010). Toward signal processing theory for graphs and non-Euclidean data. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 5414-5417). [5494930] https://doi.org/10.1109/ICASSP.2010.5494930

Toward signal processing theory for graphs and non-Euclidean data. / Miller, Benjamin A.; Bliss, Nadya; Wolfe, Patrick J.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 5414-5417 5494930.

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

Miller, BA, Bliss, N & Wolfe, PJ 2010, Toward signal processing theory for graphs and non-Euclidean data. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5494930, pp. 5414-5417, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 3/14/10. https://doi.org/10.1109/ICASSP.2010.5494930
Miller BA, Bliss N, Wolfe PJ. Toward signal processing theory for graphs and non-Euclidean data. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 5414-5417. 5494930 https://doi.org/10.1109/ICASSP.2010.5494930
Miller, Benjamin A. ; Bliss, Nadya ; Wolfe, Patrick J. / Toward signal processing theory for graphs and non-Euclidean data. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. pp. 5414-5417
@inproceedings{930c38175e6e4b77a2ea04484eb3b00c,
title = "Toward signal processing theory for graphs and non-Euclidean data",
abstract = "Graphs are canonical examples of high-dimensional non-Euclidean data sets, and are emerging as a common data structure in many fields. While there are many algorithms to analyze such data, a signal processing theory for evaluating these techniques akin to detection and estimation in the classical Euclidean setting remains to be developed. In this paper we show the conceptual advantages gained by formulating graph analysis problems in a signal processing framework by way of a practical example: detection of a subgraph embedded in a background graph. We describe an approach based on detection theory and provide empirical results indicating that the test statistic proposed has reasonable power to detect dense subgraphs in large random graphs.",
keywords = "Chi-squared test, Community detection, Graph algorithms, High-dimensional data, Signal detection theory",
author = "Miller, {Benjamin A.} and Nadya Bliss and Wolfe, {Patrick J.}",
year = "2010",
doi = "10.1109/ICASSP.2010.5494930",
language = "English (US)",
isbn = "9781424442966",
pages = "5414--5417",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Toward signal processing theory for graphs and non-Euclidean data

AU - Miller, Benjamin A.

AU - Bliss, Nadya

AU - Wolfe, Patrick J.

PY - 2010

Y1 - 2010

N2 - Graphs are canonical examples of high-dimensional non-Euclidean data sets, and are emerging as a common data structure in many fields. While there are many algorithms to analyze such data, a signal processing theory for evaluating these techniques akin to detection and estimation in the classical Euclidean setting remains to be developed. In this paper we show the conceptual advantages gained by formulating graph analysis problems in a signal processing framework by way of a practical example: detection of a subgraph embedded in a background graph. We describe an approach based on detection theory and provide empirical results indicating that the test statistic proposed has reasonable power to detect dense subgraphs in large random graphs.

AB - Graphs are canonical examples of high-dimensional non-Euclidean data sets, and are emerging as a common data structure in many fields. While there are many algorithms to analyze such data, a signal processing theory for evaluating these techniques akin to detection and estimation in the classical Euclidean setting remains to be developed. In this paper we show the conceptual advantages gained by formulating graph analysis problems in a signal processing framework by way of a practical example: detection of a subgraph embedded in a background graph. We describe an approach based on detection theory and provide empirical results indicating that the test statistic proposed has reasonable power to detect dense subgraphs in large random graphs.

KW - Chi-squared test

KW - Community detection

KW - Graph algorithms

KW - High-dimensional data

KW - Signal detection theory

UR - http://www.scopus.com/inward/record.url?scp=78049365565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049365565&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5494930

DO - 10.1109/ICASSP.2010.5494930

M3 - Conference contribution

SN - 9781424442966

SP - 5414

EP - 5417

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -