Minimierung des Auftretens von Anomalen Zuständen in einem Stochastischen Transaktionsverarbeitungssystem unter Verwendung eines Maschinell Gelernten Netzwerks

Details

Veröffentlichungs-Nr.

EP4730249

Anmeldetag

17. Oktober 2025

Veröffentlichung

22. April 2026

Rechtsraum

EP

IPC

G06Q40/04, // G06N3/08(RAI RAMPRASAD et al.)

Abstract

The disclosed embodiments relate to systems/methods which utilize machine learnt networks for efficiently proactively preventing an electronic data transaction processing system, such as an electronic trading system, from transitioning to an anomalous state which must be reverted, e.g., to unwind one or more processed transactions, or where further state changes must be restricted, e.g., by only allowing limited processing of subsequently received transactions, or prevented, e.g., the processing of subsequently received transactions is temporarily or permanently halted. Using a machine learnt network, the disclosed embodiments are able to recognize both anticipated and unanticipated anomalous transactions which would result in an anomalous state reducing occurrence thereof and minimizing having to revert, pause or halt the electronic trading system while also minimizing the impact on the performance of the electronic trading system.

Anmelder

Firma

Chicago Mercantile Exchange Inc.

Land

🇺🇸 USA

Fachgebiete