Outlier Detection in Graphs: On the Impact of Multiple Graph Models

Guilherme Oliveira Campos, Wagner Meira Jr., Arthur Zimek

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Various previous works proposed techniques to detect outliers in graph data. Usually, some complex dataset is modeled as a graph and a technique for detecting outliers in graphs is applied. The impact of the graph model on the outlier detection capabilities of any method has been ignored. Here we assess the impact of the graph model on the outlier detection performance and the gains that may be achieved by using multiple graph models and combining the results obtained by these models. We show that assessing the similarity between graphs may be a guidance to determine effective combinations, as less similar graphs are complementary with respect to outlier information they provide and lead to better outlier detection.
Original languageEnglish
Title of host publicationProceedings of the 8th International Conference on Web Intelligence, Mining and Semantics : WIMS 2018
EditorsCostin Badica, Rajendra Akerkar, Mirjana Ivanovic, Milos Savic, Milos Radovanovic, Sang-Wook Kim, Riccardo Rosati, Yannis Manolopoulos
Number of pages12
PublisherAssociation for Computing Machinery
Publication date25. Jun 2018
Article number21
ISBN (Electronic)978-1-4503-5489-9
DOIs
Publication statusPublished - 25. Jun 2018
Event8th International Conference on Web Intelligence, Mining and Semantics - Novi Sad, Serbia
Duration: 25. Jun 201827. Jun 2018
https://wims2018.pmf.uns.ac.rs/

Conference

Conference8th International Conference on Web Intelligence, Mining and Semantics
CountrySerbia
CityNovi Sad
Period25/06/201827/06/2018
Internet address

Keywords

  • Ensemble
  • Multiple graph models
  • Outlier detection

Cite this

Campos, G. O., Meira Jr., W., & Zimek, A. (2018). Outlier Detection in Graphs: On the Impact of Multiple Graph Models. In C. Badica, R. Akerkar, M. Ivanovic, M. Savic, M. Radovanovic, S-W. Kim, R. Rosati, ... Y. Manolopoulos (Eds.), Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics: WIMS 2018 [21] Association for Computing Machinery. https://doi.org/10.1145/3227609.3227646
Campos, Guilherme Oliveira ; Meira Jr., Wagner ; Zimek, Arthur. / Outlier Detection in Graphs: On the Impact of Multiple Graph Models. Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics: WIMS 2018. editor / Costin Badica ; Rajendra Akerkar ; Mirjana Ivanovic ; Milos Savic ; Milos Radovanovic ; Sang-Wook Kim ; Riccardo Rosati ; Yannis Manolopoulos. Association for Computing Machinery, 2018.
@inproceedings{e467ec927c3c4b4aa3cc8bd1ffd3876e,
title = "Outlier Detection in Graphs: On the Impact of Multiple Graph Models",
abstract = "Various previous works proposed techniques to detect outliers in graph data. Usually, some complex dataset is modeled as a graph and a technique for detecting outliers in graphs is applied. The impact of the graph model on the outlier detection capabilities of any method has been ignored. Here we assess the impact of the graph model on the outlier detection performance and the gains that may be achieved by using multiple graph models and combining the results obtained by these models. We show that assessing the similarity between graphs may be a guidance to determine effective combinations, as less similar graphs are complementary with respect to outlier information they provide and lead to better outlier detection.",
keywords = "Ensemble, Multiple graph models, Outlier detection",
author = "Campos, {Guilherme Oliveira} and {Meira Jr.}, Wagner and Arthur Zimek",
year = "2018",
month = "6",
day = "25",
doi = "10.1145/3227609.3227646",
language = "English",
editor = "Costin Badica and Rajendra Akerkar and Mirjana Ivanovic and Milos Savic and Milos Radovanovic and Sang-Wook Kim and Riccardo Rosati and Yannis Manolopoulos",
booktitle = "Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics",
publisher = "Association for Computing Machinery",
address = "United States",

}

Campos, GO, Meira Jr., W & Zimek, A 2018, Outlier Detection in Graphs: On the Impact of Multiple Graph Models. in C Badica, R Akerkar, M Ivanovic, M Savic, M Radovanovic, S-W Kim, R Rosati & Y Manolopoulos (eds), Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics: WIMS 2018., 21, Association for Computing Machinery, 8th International Conference on Web Intelligence, Mining and Semantics, Novi Sad, Serbia, 25/06/2018. https://doi.org/10.1145/3227609.3227646

Outlier Detection in Graphs: On the Impact of Multiple Graph Models. / Campos, Guilherme Oliveira; Meira Jr., Wagner; Zimek, Arthur.

Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics: WIMS 2018. ed. / Costin Badica; Rajendra Akerkar; Mirjana Ivanovic; Milos Savic; Milos Radovanovic; Sang-Wook Kim; Riccardo Rosati; Yannis Manolopoulos. Association for Computing Machinery, 2018. 21.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Outlier Detection in Graphs: On the Impact of Multiple Graph Models

AU - Campos, Guilherme Oliveira

AU - Meira Jr., Wagner

AU - Zimek, Arthur

PY - 2018/6/25

Y1 - 2018/6/25

N2 - Various previous works proposed techniques to detect outliers in graph data. Usually, some complex dataset is modeled as a graph and a technique for detecting outliers in graphs is applied. The impact of the graph model on the outlier detection capabilities of any method has been ignored. Here we assess the impact of the graph model on the outlier detection performance and the gains that may be achieved by using multiple graph models and combining the results obtained by these models. We show that assessing the similarity between graphs may be a guidance to determine effective combinations, as less similar graphs are complementary with respect to outlier information they provide and lead to better outlier detection.

AB - Various previous works proposed techniques to detect outliers in graph data. Usually, some complex dataset is modeled as a graph and a technique for detecting outliers in graphs is applied. The impact of the graph model on the outlier detection capabilities of any method has been ignored. Here we assess the impact of the graph model on the outlier detection performance and the gains that may be achieved by using multiple graph models and combining the results obtained by these models. We show that assessing the similarity between graphs may be a guidance to determine effective combinations, as less similar graphs are complementary with respect to outlier information they provide and lead to better outlier detection.

KW - Ensemble

KW - Multiple graph models

KW - Outlier detection

U2 - 10.1145/3227609.3227646

DO - 10.1145/3227609.3227646

M3 - Article in proceedings

BT - Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics

A2 - Badica, Costin

A2 - Akerkar, Rajendra

A2 - Ivanovic, Mirjana

A2 - Savic, Milos

A2 - Radovanovic, Milos

A2 - Kim, Sang-Wook

A2 - Rosati, Riccardo

A2 - Manolopoulos, Yannis

PB - Association for Computing Machinery

ER -

Campos GO, Meira Jr. W, Zimek A. Outlier Detection in Graphs: On the Impact of Multiple Graph Models. In Badica C, Akerkar R, Ivanovic M, Savic M, Radovanovic M, Kim S-W, Rosati R, Manolopoulos Y, editors, Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics: WIMS 2018. Association for Computing Machinery. 2018. 21 https://doi.org/10.1145/3227609.3227646