A news monitoring system to detect relevant news for the anti-money laundering supervision of financial institutions.

Kris Boudt; Olivier Delmarcelle; Pascal Ringoot

doi:10.1016/j.risk.2025.100018

A news monitoring system to detect relevant news for the anti-money laundering supervision of financial institutions.

Kris Boudt, Olivier Delmarcelle, Pascal Ringoot

Research output: Contribution to journal › Article › peer-review

3 Downloads (Pure)

Abstract

Investigative journalism plays a pivotal role in uncovering financial misconduct, yet its integration into anti-money laundering (AML) supervision remains underexplored. This paper presents a data-driven media monitoring system designed to provide timely alerts on news articles relevant to AML oversight in financial institutions. The system produces early-warning risk signals, offering actionable insights for qualitative assessments by micro-prudential supervisors. Relevance is quantified through thematic and entity-specific keywords, with these scores aggregated into weekly risk indicators. Through event analysis, we demonstrate the system's effectiveness by examining how Belgian newspapers covered eight offshore leaks between 2013 and 2021. Furthermore, we assess the system's robustness to machine translation and explore the use of prompting as an alternative methodology. The findings highlight the potential of integrating news-based alerts into AML supervision, enhancing the monitoring and response capabilities of supervisory authorities.

Original language	English
Number of pages	19
Journal	Risk Sciences
Volume	1
DOIs	https://doi.org/10.1016/j.risk.2025.100018
Publication status	Published - 2025

Access to Document

10.1016/j.risk.2025.100018Licence: CC BY-NC-ND

133783355Final published version, 4.18 MBLicence: CC BY-NC-ND

Handle.net

Persistent link

Cite this

@article{6541ce7a10fc4188b7bd6fd7e716bf86,

title = "A news monitoring system to detect relevant news for the anti-money laundering supervision of financial institutions.",

abstract = "Investigative journalism plays a pivotal role in uncovering financial misconduct, yet its integration into anti-money laundering (AML) supervision remains underexplored. This paper presents a data-driven media monitoring system designed to provide timely alerts on news articles relevant to AML oversight in financial institutions. The system produces early-warning risk signals, offering actionable insights for qualitative assessments by micro-prudential supervisors. Relevance is quantified through thematic and entity-specific keywords, with these scores aggregated into weekly risk indicators. Through event analysis, we demonstrate the system's effectiveness by examining how Belgian newspapers covered eight offshore leaks between 2013 and 2021. Furthermore, we assess the system's robustness to machine translation and explore the use of prompting as an alternative methodology. The findings highlight the potential of integrating news-based alerts into AML supervision, enhancing the monitoring and response capabilities of supervisory authorities.",

author = "Kris Boudt and Olivier Delmarcelle and Pascal Ringoot",

year = "2025",

doi = "10.1016/j.risk.2025.100018",

language = "English",

volume = "1",

journal = "Risk Sciences",

issn = "2950-6298",

publisher = "Beijing: Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.",

}

TY - JOUR

T1 - A news monitoring system to detect relevant news for the anti-money laundering supervision of financial institutions.

AU - Boudt, Kris

AU - Delmarcelle, Olivier

AU - Ringoot, Pascal

PY - 2025

Y1 - 2025

N2 - Investigative journalism plays a pivotal role in uncovering financial misconduct, yet its integration into anti-money laundering (AML) supervision remains underexplored. This paper presents a data-driven media monitoring system designed to provide timely alerts on news articles relevant to AML oversight in financial institutions. The system produces early-warning risk signals, offering actionable insights for qualitative assessments by micro-prudential supervisors. Relevance is quantified through thematic and entity-specific keywords, with these scores aggregated into weekly risk indicators. Through event analysis, we demonstrate the system's effectiveness by examining how Belgian newspapers covered eight offshore leaks between 2013 and 2021. Furthermore, we assess the system's robustness to machine translation and explore the use of prompting as an alternative methodology. The findings highlight the potential of integrating news-based alerts into AML supervision, enhancing the monitoring and response capabilities of supervisory authorities.

AB - Investigative journalism plays a pivotal role in uncovering financial misconduct, yet its integration into anti-money laundering (AML) supervision remains underexplored. This paper presents a data-driven media monitoring system designed to provide timely alerts on news articles relevant to AML oversight in financial institutions. The system produces early-warning risk signals, offering actionable insights for qualitative assessments by micro-prudential supervisors. Relevance is quantified through thematic and entity-specific keywords, with these scores aggregated into weekly risk indicators. Through event analysis, we demonstrate the system's effectiveness by examining how Belgian newspapers covered eight offshore leaks between 2013 and 2021. Furthermore, we assess the system's robustness to machine translation and explore the use of prompting as an alternative methodology. The findings highlight the potential of integrating news-based alerts into AML supervision, enhancing the monitoring and response capabilities of supervisory authorities.

U2 - 10.1016/j.risk.2025.100018

DO - 10.1016/j.risk.2025.100018

M3 - Article

SN - 2950-6298

VL - 1

JO - Risk Sciences

JF - Risk Sciences

ER -