Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing

Corentin Larroche; Johan Mazel; Stephan Clémençon

doi:10.14428/esann/2021.ES2021-32

Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing

Corentin Larroche
, Johan Mazel
, Stephan Clémençon

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

Abstract

We propose a computationally efficient procedure for elevated mean detection on a connected subgraph of a network with node-related scalar observations. Our approach relies on two intuitions: first, a significant concentration of high observations in a connected subgraph implies that the subgraph induced by the nodes associated with the highest observations has a large connected component. Secondly, a greater detection power can be obtained in certain cases by denoising the observations using the network structure. Numerical experiments show that our procedure's detection performance and computational efficiency are both competitive.

Original language	English
Title of host publication	ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Publisher	i6doc.com publication
Pages	399-404
Number of pages	6
ISBN (Electronic)	9782875870827
DOIs	https://doi.org/10.14428/esann/2021.ES2021-32
Publication status	Published - 1 Jan 2021
Event	29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021 - Virtual, Online, Belgium Duration: 6 Oct 2021 → 8 Oct 2021

Publication series

Name	ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Conference

Conference	29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021
Country/Territory	Belgium
City	Virtual, Online
Period	6/10/21 → 8/10/21

Access to Document

10.14428/esann/2021.ES2021-32

Cite this

Larroche, C., Mazel, J., & Clémençon, S. (2021). Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing. In ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (pp. 399-404). (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning). i6doc.com publication. https://doi.org/10.14428/esann/2021.ES2021-32

Larroche, Corentin ; Mazel, Johan ; Clémençon, Stephan. / Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing. ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication, 2021. pp. 399-404 (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).

@inproceedings{5e414a3b1314411ab5af27475738d622,

title = "Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing",

abstract = "We propose a computationally efficient procedure for elevated mean detection on a connected subgraph of a network with node-related scalar observations. Our approach relies on two intuitions: first, a significant concentration of high observations in a connected subgraph implies that the subgraph induced by the nodes associated with the highest observations has a large connected component. Secondly, a greater detection power can be obtained in certain cases by denoising the observations using the network structure. Numerical experiments show that our procedure's detection performance and computational efficiency are both competitive.",

author = "Corentin Larroche and Johan Mazel and Stephan Cl{\'e}men{\c c}on",

note = "Publisher Copyright: {\textcopyright} 2021 ESANN Intelligence and Machine Learning. All rights reserved.; 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021 ; Conference date: 06-10-2021 Through 08-10-2021",

year = "2021",

month = jan,

day = "1",

doi = "10.14428/esann/2021.ES2021-32",

language = "English",

series = "ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

publisher = "i6doc.com publication",

pages = "399--404",

booktitle = "ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

}

Larroche, C, Mazel, J & Clémençon, S 2021, Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing. in ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, i6doc.com publication, pp. 399-404, 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021, Virtual, Online, Belgium, 6/10/21. https://doi.org/10.14428/esann/2021.ES2021-32

Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing. / Larroche, Corentin; Mazel, Johan; Clémençon, Stephan.
ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication, 2021. p. 399-404 (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).

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

TY - GEN

T1 - Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing

AU - Larroche, Corentin

AU - Mazel, Johan

AU - Clémençon, Stephan

PY - 2021/1/1

Y1 - 2021/1/1

N2 - We propose a computationally efficient procedure for elevated mean detection on a connected subgraph of a network with node-related scalar observations. Our approach relies on two intuitions: first, a significant concentration of high observations in a connected subgraph implies that the subgraph induced by the nodes associated with the highest observations has a large connected component. Secondly, a greater detection power can be obtained in certain cases by denoising the observations using the network structure. Numerical experiments show that our procedure's detection performance and computational efficiency are both competitive.

AB - We propose a computationally efficient procedure for elevated mean detection on a connected subgraph of a network with node-related scalar observations. Our approach relies on two intuitions: first, a significant concentration of high observations in a connected subgraph implies that the subgraph induced by the nodes associated with the highest observations has a large connected component. Secondly, a greater detection power can be obtained in certain cases by denoising the observations using the network structure. Numerical experiments show that our procedure's detection performance and computational efficiency are both competitive.

U2 - 10.14428/esann/2021.ES2021-32

DO - 10.14428/esann/2021.ES2021-32

M3 - Conference contribution

AN - SCOPUS:85129239833

T3 - ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

SP - 399

EP - 404

BT - ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

PB - i6doc.com publication

T2 - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021

Y2 - 6 October 2021 through 8 October 2021

ER -

Larroche C, Mazel J, Clémençon S. Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing. In ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication. 2021. p. 399-404. (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning). doi: 10.14428/esann/2021.ES2021-32

Anomalous Cluster Detection in Large Networks with Diffusion-Percolation Testing

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this