Rulesets including optimized parameters for detecting fraud
GB2632150APending Publication Date: 2025-01-29SOPRA STERIA LTD
Patent Information
- Application Number
- GB2023011491
- Authority / Receiving Office
- GB · GB
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-07-26
- Publication Date
- 2025-01-29
Smart Images
Figure 00000000_0000_ABST
Abstract
System and method for creating a plurality of rulesets 164 for a financial institution includes receiving details of a plurality of transactions 110 and analysing the details using a trained machine learning model 120 configured to determine if at least one of the plurality of transactions is fraudulent. The details of a transactions are verified based on the analysis and the plurality of rulesets are generated based on the verified details. The plurality of rulesets includes a plurality of parameters 162 based on the details of the transactions. The rulesets configured to be used in the financial institution are then delivered. The parameters may be at least one of a transaction amount, date of the transaction, method of transaction, a model score of the transaction, or frequency of the transaction. The verifying step may comprise determining if the at least one of the plurality of transactions is declined or permitted.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Machine learning engine for fraud detection during cross-location online transaction processing
US20200005310A1
Tuning fraud-detection rules using machine learning
US20200334679A1
Identification and Suggestion of Rules Using Machine Learning
US20200387835A1
Methods, systems, and apparatuses for improved fraud detection and reduction
US20210042824A1
Systems and methods for improved fraud detection
US20210081948A1
Cited By
Systems and methods for assessing fraud risk using machine learning
US20250245666A1