Fraud Detection Model Ranking to Eliminate Duplicate Alerts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fraud detection systems face inefficiencies due to overlapping and duplicate fraud alerts, leading to resource wastage and difficulty in determining the net contribution of each fraud detection rule or model, which complicates the prompt identification of fraudulent transactions.
Innovation Solution
A computing device, the fraud optimizer device, employs a customized forward selection algorithm to evaluate and rank fraud detection rules/models by calculating a fraud value for each alert, removing duplicates, and creating a ranked list to optimize resource usage and minimize overlaps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple fraud detection strategies are applied to detect fraudulent transactions, then the detection coverage is improved, but the computational resources and processing time increase due to overlapping and duplicate alerts
Solution Approach 1:
The patent segments the fraud detection process by applying multiple independent fraud detection strategies to different aspects of transaction analysis. Each strategy operates separately to identify potential fraud indicators, and the results are then consolidated by removing duplicates. This segmentation allows comprehensive coverage while managing computational load through structured organization of detection efforts.
Solution Approach 2:
The patent merges results from multiple fraud detection strategies by consolidating their outputs and removing duplicate alerts. The system combines the detection capabilities of various strategies while eliminating redundant processing of the same transactions, thereby maintaining high detection coverage without proportionally increasing computational resources.
2Reliability
If all fraud detection strategies are processed equally, then comprehensive fraud detection is achieved, but resource allocation becomes inefficient due to inability to prioritize high-value alerts
Solution Approach 1:
The patent changes the parameter of alert evaluation by introducing a fraud value metric that quantifies the significance of each alert. This parameter transformation allows the system to differentiate between high-value and low-value alerts, enabling prioritized processing that reduces computational resource waste while maintaining comprehensive detection coverage.
Solution Approach 2:
The system automatically evaluates and ranks fraud alerts based on their calculated fraud values without requiring manual intervention. The fraud detection model self-optimizes by identifying and prioritizing high-value alerts, allocating computational resources efficiently to the most significant detection tasks.
3Measurement precision
If duplicate transactions are processed by multiple strategies, then detection accuracy is improved through multiple perspectives, but processing time increases due to redundant analysis
Solution Approach 1:
The patent applies preliminary action by removing duplicate transactions from the processing queue before they can be analyzed by multiple strategies. The system identifies and eliminates redundant alerts upfront, ensuring that each unique transaction is processed only once while still benefiting from multiple detection perspectives through the initial comprehensive scanning phase.
Data Source
AI summary
There is provided a computing system for optimizing a plurality of fraud detection strategies used to generate a corresponding set of potentially fraudulent transactions. The system determines an overall fraud value such as an average fraud value for each transaction based on pre-defined factors and identifies a particular strategy having a highest average fraud value for its fraudulent transactions as a highest priority on a ranked list of strategies. The system is configured to remove each transaction from the remaining other strategies if the same as the fraudulent transactions in the identified strategy and calculate an average fraud value for the remaining other strategies. The system then ranks the next highest priority fraud detection strategy having the highest average fraud value while removing its corresponding transactions flagged from other remaining strategies and repeat the ranking until all the strategies have been ranked and apply the ranked list to subsequent transactions.