Anomaly Detection in Real-Time Gross Settlement Systems

Ron Triepels; Hennie Daniels; Ronald Heijmans

doi:10.5220/0006333004330441

Anomaly Detection in Real-Time Gross Settlement Systems

Ron Triepels^*, Hennie Daniels, Ronald Heijmans

^*Corresponding author for this work

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

Abstract

We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.

Original language	English
Title of host publication	Proceedings of the 19th International Conference on Enterprise Information Systems
Place of Publication	Porto, Portugal
Publisher	Scitepress - Science And Technology Publications
Pages	433-441
Volume	3
ISBN (Print)	978-989-758-247-9
DOIs	https://doi.org/10.5220/0006333004330441
Publication status	Published - 2017
Externally published	Yes
Event	19th International Conference on Enterprise Information Systems - Porto, Portugal Duration: 26 Apr 2017 → 29 Apr 2017 http://www.iceis.org/?y=2017

Conference

Conference	19th International Conference on Enterprise Information Systems
Abbreviated title	ICEIS 2017
Country/Territory	Portugal
City	Porto
Period	26/04/17 → 29/04/17
Internet address	http://www.iceis.org/?y=2017

Keywords

Anomaly detection
Neural network
Autoencoder
Real-Time gross settlement system

Access to Document

10.5220/0006333004330441Licence: CC BY-NC-ND

Cite this

@inproceedings{be4c4fabd1534073a112fec3557bb837,

title = "Anomaly Detection in Real-Time Gross Settlement Systems",

abstract = "We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.",

keywords = "Anomaly detection, Neural network, Autoencoder, Real-Time gross settlement system",

author = "Ron Triepels and Hennie Daniels and Ronald Heijmans",

year = "2017",

doi = "10.5220/0006333004330441",

language = "English",

isbn = "978-989-758-247-9",

volume = "3",

pages = "433--441",

booktitle = "Proceedings of the 19th International Conference on Enterprise Information Systems",

publisher = "Scitepress - Science And Technology Publications",

address = "Portugal",

note = "19th International Conference on Enterprise Information Systems, ICEIS 2017 ; Conference date: 26-04-2017 Through 29-04-2017",

url = "http://www.iceis.org/?y=2017",

}

Triepels, R, Daniels, H & Heijmans, R 2017, Anomaly Detection in Real-Time Gross Settlement Systems. in Proceedings of the 19th International Conference on Enterprise Information Systems. vol. 3, Scitepress - Science And Technology Publications, Porto, Portugal, pp. 433-441, 19th International Conference on Enterprise Information Systems, Porto, Portugal, 26/04/17. https://doi.org/10.5220/0006333004330441

Anomaly Detection in Real-Time Gross Settlement Systems. / Triepels, Ron; Daniels, Hennie; Heijmans, Ronald.
Proceedings of the 19th International Conference on Enterprise Information Systems. Vol. 3 Porto, Portugal: Scitepress - Science And Technology Publications, 2017. p. 433-441.

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

TY - GEN

T1 - Anomaly Detection in Real-Time Gross Settlement Systems

AU - Triepels, Ron

AU - Daniels, Hennie

AU - Heijmans, Ronald

PY - 2017

Y1 - 2017

N2 - We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.

AB - We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.

KW - Anomaly detection

KW - Neural network

KW - Autoencoder

KW - Real-Time gross settlement system

U2 - 10.5220/0006333004330441

DO - 10.5220/0006333004330441

M3 - Conference article in proceeding

SN - 978-989-758-247-9

VL - 3

SP - 433

EP - 441

BT - Proceedings of the 19th International Conference on Enterprise Information Systems

PB - Scitepress - Science And Technology Publications

CY - Porto, Portugal

T2 - 19th International Conference on Enterprise Information Systems

Y2 - 26 April 2017 through 29 April 2017

ER -

Anomaly Detection in Real-Time Gross Settlement Systems

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this