Graph-Based Fraud Detection with the Free Energy Distance

Sylvain Courtain; Bertrand Lebichot; Ilkka Kivimäki; Marco Saerens

doi:10.1007/978-3-030-36683-4_4

Graph-Based Fraud Detection with the Free Energy Distance

Sylvain Courtain^*
, Bertrand Lebichot
, Ilkka Kivimäki
, Marco Saerens

^*Corresponding author for this work

Department of Computer Science

Université Catholique de Louvain
Université libre de Bruxelles

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

6 Citations (Scopus)

Abstract

This paper investigates a real-world application of the free energy distance between nodes of a graph [14, 20] by proposing an improved extension of the existing Fraud Detection System named APATE [36]. It relies on a new way of computing the free energy distance based on paths of increasing length, and scaling on large, sparse, graphs. This new approach is assessed on a real-world large-scale e-commerce payment transactions dataset obtained from a major Belgian credit card issuer. Our results show that the free-energy based approach reduces the computation time by one half while maintaining state-of-the art performance in term of Precision@100 on fraudulent card prediction.

Original language	English
Title of host publication	Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019
Editors	Hocine Cherifi, Sabrina Gaito, José Fernendo Mendes, Esteban Moro, Luis Mateus Rocha
Publisher	Springer
Pages	40-52
Number of pages	13
ISBN (Print)	9783030366827
DOIs	https://doi.org/10.1007/978-3-030-36683-4_4
Publication status	Published - 2020
MoE publication type	A4 Conference publication
Event	International Conference on Complex Networks and their Applications - Lisbon, Portugal Duration: 10 Dec 2019 → 12 Dec 2019 Conference number: 8 https://www.complexnetworks.org/

Publication series

Name	Studies in Computational Intelligence
Volume	882 SCI
ISSN (Print)	1860-949X
ISSN (Electronic)	1860-9503

Conference

Conference	International Conference on Complex Networks and their Applications
Abbreviated title	Complex Networks
Country/Territory	Portugal
City	Lisbon
Period	10/12/2019 → 12/12/2019
Internet address	https://www.complexnetworks.org/

Funding

This work was partially supported by the Immediate funded by Wallon Region project and by the Defeatfrauds project funded by Innoviris. We thank these institutions for giving us the opportunity to conduct both fundamental and applied research. We also thank Worldline SA/NV, R&D, for providing us the data and expertise.

Keywords

Credit card fraud detection
Free energy distance
Network data analysis
Network science
Semi-supervised learning

Access to Document

10.1007/978-3-030-36683-4_4

Cite this

Courtain, S., Lebichot, B., Kivimäki, I., & Saerens, M. (2020). Graph-Based Fraud Detection with the Free Energy Distance. In H. Cherifi, S. Gaito, J. F. Mendes, E. Moro, & L. M. Rocha (Eds.), Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019 (pp. 40-52). (Studies in Computational Intelligence; Vol. 882 SCI). Springer. https://doi.org/10.1007/978-3-030-36683-4_4

Courtain, Sylvain ; Lebichot, Bertrand ; Kivimäki, Ilkka et al. / Graph-Based Fraud Detection with the Free Energy Distance. Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019. editor / Hocine Cherifi ; Sabrina Gaito ; José Fernendo Mendes ; Esteban Moro ; Luis Mateus Rocha. Springer, 2020. pp. 40-52 (Studies in Computational Intelligence).

@inproceedings{a2b0d4e5c34c4530bae396349a22e230,

title = "Graph-Based Fraud Detection with the Free Energy Distance",

abstract = "This paper investigates a real-world application of the free energy distance between nodes of a graph [14, 20] by proposing an improved extension of the existing Fraud Detection System named APATE [36]. It relies on a new way of computing the free energy distance based on paths of increasing length, and scaling on large, sparse, graphs. This new approach is assessed on a real-world large-scale e-commerce payment transactions dataset obtained from a major Belgian credit card issuer. Our results show that the free-energy based approach reduces the computation time by one half while maintaining state-of-the art performance in term of Precision@100 on fraudulent card prediction.",

keywords = "Credit card fraud detection, Free energy distance, Network data analysis, Network science, Semi-supervised learning",

author = "Sylvain Courtain and Bertrand Lebichot and Ilkka Kivim{\"a}ki and Marco Saerens",

year = "2020",

doi = "10.1007/978-3-030-36683-4\_4",

language = "English",

isbn = "9783030366827",

series = "Studies in Computational Intelligence",

publisher = "Springer",

pages = "40--52",

editor = "Hocine Cherifi and Sabrina Gaito and Mendes, \{Jos{\'e} Fernendo\} and Esteban Moro and Rocha, \{Luis Mateus\}",

booktitle = "Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019",

address = "Germany",

note = "International Conference on Complex Networks and their Applications, Complex Networks ; Conference date: 10-12-2019 Through 12-12-2019",

url = "https://www.complexnetworks.org/",

}

Courtain, S, Lebichot, B, Kivimäki, I & Saerens, M 2020, Graph-Based Fraud Detection with the Free Energy Distance. in H Cherifi, S Gaito, JF Mendes, E Moro & LM Rocha (eds), Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019. Studies in Computational Intelligence, vol. 882 SCI, Springer, pp. 40-52, International Conference on Complex Networks and their Applications, Lisbon, Portugal, 10/12/2019. https://doi.org/10.1007/978-3-030-36683-4_4

Graph-Based Fraud Detection with the Free Energy Distance. / Courtain, Sylvain; Lebichot, Bertrand; Kivimäki, Ilkka et al.
Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019. ed. / Hocine Cherifi; Sabrina Gaito; José Fernendo Mendes; Esteban Moro; Luis Mateus Rocha. Springer, 2020. p. 40-52 (Studies in Computational Intelligence; Vol. 882 SCI).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Graph-Based Fraud Detection with the Free Energy Distance

AU - Courtain, Sylvain

AU - Lebichot, Bertrand

AU - Kivimäki, Ilkka

AU - Saerens, Marco

N1 - Conference code: 8

PY - 2020

Y1 - 2020

N2 - This paper investigates a real-world application of the free energy distance between nodes of a graph [14, 20] by proposing an improved extension of the existing Fraud Detection System named APATE [36]. It relies on a new way of computing the free energy distance based on paths of increasing length, and scaling on large, sparse, graphs. This new approach is assessed on a real-world large-scale e-commerce payment transactions dataset obtained from a major Belgian credit card issuer. Our results show that the free-energy based approach reduces the computation time by one half while maintaining state-of-the art performance in term of Precision@100 on fraudulent card prediction.

AB - This paper investigates a real-world application of the free energy distance between nodes of a graph [14, 20] by proposing an improved extension of the existing Fraud Detection System named APATE [36]. It relies on a new way of computing the free energy distance based on paths of increasing length, and scaling on large, sparse, graphs. This new approach is assessed on a real-world large-scale e-commerce payment transactions dataset obtained from a major Belgian credit card issuer. Our results show that the free-energy based approach reduces the computation time by one half while maintaining state-of-the art performance in term of Precision@100 on fraudulent card prediction.

KW - Credit card fraud detection

KW - Free energy distance

KW - Network data analysis

KW - Network science

KW - Semi-supervised learning

UR - https://www.scopus.com/pages/publications/85087916148

U2 - 10.1007/978-3-030-36683-4_4

DO - 10.1007/978-3-030-36683-4_4

M3 - Conference article in proceedings

AN - SCOPUS:85087916148

SN - 9783030366827

T3 - Studies in Computational Intelligence

SP - 40

EP - 52

BT - Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019

A2 - Cherifi, Hocine

A2 - Gaito, Sabrina

A2 - Mendes, José Fernendo

A2 - Moro, Esteban

A2 - Rocha, Luis Mateus

PB - Springer

T2 - International Conference on Complex Networks and their Applications

Y2 - 10 December 2019 through 12 December 2019

ER -

Courtain S, Lebichot B, Kivimäki I, Saerens M. Graph-Based Fraud Detection with the Free Energy Distance. In Cherifi H, Gaito S, Mendes JF, Moro E, Rocha LM, editors, Complex Networks and Their Applications VIII - Volume 2 Proceedings of the 8th International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2019. Springer. 2020. p. 40-52. (Studies in Computational Intelligence). doi: 10.1007/978-3-030-36683-4_4

Graph-Based Fraud Detection with the Free Energy Distance

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this