Fuzzy Inference for Financial Account Risk Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial institutions face challenges in accurately identifying and ranking accounts involved in illegal activities, such as money laundering, due to high volumes of false positive alerts generated by traditional risk assessment methods, which can overwhelm compliance officers and increase resource costs.
Innovation Solution
A fuzzy inference-based system that combines Bayesian inference with adaptive thresholds and fuzzy set theory to assess risk scores, allowing for dynamic and account-specific evaluation of transaction patterns, reducing false positives by considering historical behavior and specific account usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional risk assessment rules with fixed thresholds are used to monitor account activity, then illegal activities can be detected, but high volumes of false positive alerts are generated that overwhelm compliance officers
Solution Approach 1:
The patent transforms fixed threshold parameters into adaptive, account-specific thresholds that dynamically adjust based on historical transaction patterns. This allows the system to maintain high detection accuracy while reducing false positives by evaluating transactions relative to each account's normal behavior rather than against static rules.
Solution Approach 2:
The system performs preliminary analysis by establishing baseline transaction patterns for each account before evaluating new transactions. This pre-computation of account-specific norms enables the system to quickly assess whether new transactions deviate from expected behavior, reducing the need for manual review of normal transactions.
2Reliability
If traditional risk assessment methods are used, then suspicious activities can be identified, but resource costs increase due to high volumes of alerts requiring manual review
Solution Approach 1:
By changing from fixed thresholds to adaptive, account-specific parameters, the system maintains reliable suspicious activity identification while significantly reducing the volume of alerts requiring manual review, thereby lowering resource costs.
Solution Approach 2:
The system enables accounts to effectively monitor themselves by establishing and comparing transactions against their own historical patterns. This self-service approach reduces the burden on compliance officers to manually evaluate every alert, as the system automatically filters out transactions consistent with normal account behavior.
3Ease of manufacture
If fixed threshold rules are applied to all accounts, then monitoring can be standardized, but account-specific usage patterns are not considered leading to false positives
Solution Approach 1:
The patent applies local quality by transitioning from uniform fixed thresholds to account-specific adaptive thresholds. Each account receives customized monitoring parameters based on its unique transaction patterns, improving measurement precision while maintaining the standardized automated monitoring framework.
Solution Approach 2:
The system introduces dynamics by making thresholds adaptive rather than static. Account-specific thresholds automatically adjust based on evolving transaction patterns, allowing the monitoring system to remain standardized in structure while being precise and flexible in application.
Data Source
AI summary
A method includes receiving an alert representing that application of at least one rule to a scenario occurring with a financial account indicates that the scenario is consistent with illegal activity. The method includes, in response to the alert, using a fuzzy inference derived from activity of the financial account to determine a likelihood that the scenario is associated with illegal activity.