Cross-Institution Fraud Detection via Centralized Transaction Mediator
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic payment transaction systems are inefficient in detecting fraud across multiple transaction accounts issued by different institutions due to incomplete transaction history and privacy regulations, leading to potential fraud undetection and increased network communications and processing time.
Innovation Solution
A processing system that identifies transaction accounts issued to the same consumer across different institutions using biographical data, tracks transactions, and determines fraud likelihood based on past transaction history, allowing for efficient fraud detection and reduced network communications by declining suspicious transactions on behalf of the issuing institutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If issuing institutions share transaction information across multiple institutions, then fraud detection capability is improved, but consumer privacy protection deteriorates
Solution Approach 1:
A central processing server acts as an intermediary between multiple issuing institutions. The server receives transaction requests, queries a centralized database containing transaction histories from multiple institutions, and returns authorization decisions. This mediator enables cross-institution fraud detection without requiring direct information sharing between institutions, thus protecting consumer privacy while improving fraud detection capability.
2Reliability
If issuing institutions access transaction history from multiple institutions, then fraud detection accuracy is improved, but network communication overhead increases
Solution Approach 1:
Transaction histories and account information from multiple issuing institutions are pre-loaded into a centralized database before fraud detection is needed. When a transaction request arrives, the processing server performs a quick query of this pre-populated database rather than initiating real-time communications with multiple institutions. This preliminary preparation significantly reduces network communication overhead while maintaining high fraud detection accuracy.
3Reliability
If issuing institutions analyze comprehensive transaction history, then fraud detection effectiveness is improved, but processing time increases
Solution Approach 1:
Complex fraud detection algorithms and analysis routines are implemented as software on the processing server rather than requiring extensive manual review processes. The server automatically compares transaction patterns against historical data, identifies anomalies, and makes authorization decisions using programmed logic. This substitution of automated computational mechanics for manual analysis dramatically reduces processing time while maintaining or improving fraud detection effectiveness.
Data Source
AI summary
A method for processing an electronic payment transaction includes: storing an account profile including a plurality of account numbers; receiving a first authorization request for a payment transaction that includes a first account number of the plurality of account numbers, a first time, and a first location; processing the first payment transaction; receiving a second authorization request for a different transaction including a second primary account number of the plurality of account numbers, a second time, and a second location, where the second account number is associated with a different issuer than the first account number; determining a likelihood of fraud based on the first time and location and the second time and location; and forwarding the second authorization request to the different issuer if the likelihood of fraud is below a threshold, or transmitting an authorization response indicating decline if the likelihood of fraud is above the threshold.