Multi-stage fraud filtering with velocity checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fraud detection systems are inefficient as they treat all transactions equally, leading to both missed fraudulent transactions and false-positive identifications, resulting in increased costs and consumer frustration.
Innovation Solution
A multi-stage filtering system for fraud detection that evaluates financial transactions based on progressively more detailed attributes, including velocity data, customer history, and geo-positioning, to differentiate between legitimate and potentially fraudulent transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fraud detection systems evaluate all transactions with the same strict criteria, then fraudulent transactions are identified, but false-positive identifications increase and legitimate transactions are declined
Solution Approach 1:
The fraud detection system is divided into multiple stages with increasing scrutiny. Stage 1 applies basic filters to all transactions, Stage 2 applies velocity checks to transactions that passed Stage 1, and Stage 3 applies comprehensive analysis only to transactions that passed Stage 2. This segmentation allows the system to maintain high fraud detection accuracy while reducing false positives by not applying strict criteria uniformly to all transactions.
2Reliability
If conventional fraud detection systems apply strict filtering criteria to eliminate fraudulent transactions, then fraud identification improves, but data storage costs increase and processing efficiency decreases
Solution Approach 1:
The system segments transactions into different processing paths based on risk level. Low-risk transactions are processed quickly through Stage 1 and approved, while only transactions that trigger alerts proceed to subsequent stages. This maintains high fraud identification accuracy for suspicious transactions while preserving overall processing efficiency.
Solution Approach 2:
The system applies comprehensive fraud analysis only partially - specifically to transactions that pass through Stage 1 filtering and trigger velocity alerts. Most transactions receive minimal processing, while only a small subset undergoes the full multi-stage analysis, thereby maintaining productivity while ensuring thorough scrutiny where needed.
3Measurement precision
If conventional fraud detection systems analyze each transaction in detail, then fraud detection accuracy improves, but processing time increases and system complexity increases
Solution Approach 1:
The analysis is segmented into three stages with progressively deeper scrutiny. Stage 1 uses simple attribute checks for all transactions, Stage 2 adds velocity data analysis for transactions that passed Stage 1, and Stage 3 performs detailed analysis only on transactions that passed Stage 2. This segmentation achieves high detection precision for suspicious transactions while minimizing processing time for the majority of legitimate transactions.
4Measurement precision
If conventional fraud detection systems use multiple filtering stages with velocity data, then false positives are reduced, but device complexity increases
Solution Approach 1:
The system is segmented into three distinct filtering stages, each with specific attributes evaluated. Stage 1 evaluates basic transaction attributes, Stage 2 evaluates velocity data for transactions passing Stage 1, and Stage 3 performs final analysis on transactions passing Stage 2. This segmentation reduces false positives through progressive filtering while managing complexity by organizing the system into modular, sequential stages with clear boundaries.
Data Source
AI summary
A multi-stage filtering process and system for fraud detection is disclosed. The process includes one or more preliminary filtration stages followed by one or more additional filtration stages that may include velocity filters that provide for enhanced screening for fraudulent activity. Over a plurality of transactions, a portion of the transactions are cleared for processing (e.g., deemed not likely fraudulent or of too low value to continue processing) after each filtration stage. As such, acceptable transactions are not unnecessarily scrutinized.