From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors

Venus Haghighi

doi:10.1145/3539597.3572989

From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors

Venus Haghighi^*

^*Corresponding author for this work

School of Computing

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

3 Citations (Scopus)

Abstract

Graph neural networks (GNNs), which extend deep learning models to graph-structured data, have achieved great success in many applications such as detecting malicious activities. However, GNN-based models are vulnerable to camouflage behavior of malicious actors, i.e., the performance of existing GNN-based models has been hindered significantly. In this research proposal, we follow two research directions to address this challenge. One direction focuses on enhancing the existing GNN-based models and enabling them to identify both camouflaged and non-camouflaged malicious actors. In this regard, we propose to explore an adaptive aggregation strategy, which empowers GNN-based models to handle camouflage behavior of fraudsters. The other research direction concentrates on leveraging hypergraph neural networks (hyper-GNNs) to learn nodes' representation for more effective identification of camouflaged malicious actors.

Original language	English
Title of host publication	WSDM ’23
Subtitle of host publication	proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining
Place of Publication	New York
Publisher	Association for Computing Machinery
Pages	1220-1221
Number of pages	2
Volume	1
ISBN (Electronic)	9781450394079
DOIs	https://doi.org/10.1145/3539597.3572989
Publication status	Published - 2023
Event	16th ACM International Conference on Web Search and Data Mining, WSDM 2023 - Singapore, Singapore Duration: 27 Feb 2023 → 3 Mar 2023

Conference

Conference	16th ACM International Conference on Web Search and Data Mining, WSDM 2023
Country/Territory	Singapore
City	Singapore
Period	27/02/23 → 3/03/23

Keywords

Hypergraph neural networks
Graph neural networks
Camouflaged malicious actors
Homophily
Heterophily

Access to Document

10.1145/3539597.3572989

Cite this

@inbook{f07e5c41fdb1443a8d590d2e49c5e791,

title = "From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors",

abstract = "Graph neural networks (GNNs), which extend deep learning models to graph-structured data, have achieved great success in many applications such as detecting malicious activities. However, GNN-based models are vulnerable to camouflage behavior of malicious actors, i.e., the performance of existing GNN-based models has been hindered significantly. In this research proposal, we follow two research directions to address this challenge. One direction focuses on enhancing the existing GNN-based models and enabling them to identify both camouflaged and non-camouflaged malicious actors. In this regard, we propose to explore an adaptive aggregation strategy, which empowers GNN-based models to handle camouflage behavior of fraudsters. The other research direction concentrates on leveraging hypergraph neural networks (hyper-GNNs) to learn nodes' representation for more effective identification of camouflaged malicious actors.",

keywords = "Hypergraph neural networks, Graph neural networks, Camouflaged malicious actors, Homophily, Heterophily",

author = "Venus Haghighi",

year = "2023",

doi = "10.1145/3539597.3572989",

language = "English",

volume = "1",

pages = "1220--1221",

booktitle = "WSDM {\textquoteright}23",

publisher = "Association for Computing Machinery",

note = "16th ACM International Conference on Web Search and Data Mining, WSDM 2023 ; Conference date: 27-02-2023 Through 03-03-2023",

}

Haghighi, V 2023, From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors. in WSDM ’23: proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining. vol. 1, Association for Computing Machinery, New York, pp. 1220-1221, 16th ACM International Conference on Web Search and Data Mining, WSDM 2023, Singapore, Singapore, 27/02/23. https://doi.org/10.1145/3539597.3572989

From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors. / Haghighi, Venus.
WSDM ’23: proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining. Vol. 1 New York: Association for Computing Machinery, 2023. p. 1220-1221.

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

TY - CHAP

T1 - From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors

AU - Haghighi, Venus

PY - 2023

Y1 - 2023

N2 - Graph neural networks (GNNs), which extend deep learning models to graph-structured data, have achieved great success in many applications such as detecting malicious activities. However, GNN-based models are vulnerable to camouflage behavior of malicious actors, i.e., the performance of existing GNN-based models has been hindered significantly. In this research proposal, we follow two research directions to address this challenge. One direction focuses on enhancing the existing GNN-based models and enabling them to identify both camouflaged and non-camouflaged malicious actors. In this regard, we propose to explore an adaptive aggregation strategy, which empowers GNN-based models to handle camouflage behavior of fraudsters. The other research direction concentrates on leveraging hypergraph neural networks (hyper-GNNs) to learn nodes' representation for more effective identification of camouflaged malicious actors.

AB - Graph neural networks (GNNs), which extend deep learning models to graph-structured data, have achieved great success in many applications such as detecting malicious activities. However, GNN-based models are vulnerable to camouflage behavior of malicious actors, i.e., the performance of existing GNN-based models has been hindered significantly. In this research proposal, we follow two research directions to address this challenge. One direction focuses on enhancing the existing GNN-based models and enabling them to identify both camouflaged and non-camouflaged malicious actors. In this regard, we propose to explore an adaptive aggregation strategy, which empowers GNN-based models to handle camouflage behavior of fraudsters. The other research direction concentrates on leveraging hypergraph neural networks (hyper-GNNs) to learn nodes' representation for more effective identification of camouflaged malicious actors.

KW - Hypergraph neural networks

KW - Graph neural networks

KW - Camouflaged malicious actors

KW - Homophily

KW - Heterophily

UR - https://www.scopus.com/pages/publications/85149679897

U2 - 10.1145/3539597.3572989

DO - 10.1145/3539597.3572989

M3 - Conference abstract

AN - SCOPUS:85149679897

VL - 1

SP - 1220

EP - 1221

BT - WSDM ’23

PB - Association for Computing Machinery

CY - New York

T2 - 16th ACM International Conference on Web Search and Data Mining, WSDM 2023

Y2 - 27 February 2023 through 3 March 2023

ER -

From classic GNNs to hyper-GNNs for detecting camouflaged malicious actors

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this