Aggregated Payment Fraud Detection Using Household Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current transaction fraud systems monitor card activity at the individual card level, leading to limited views of spending patterns and potential missed fraudulent activities, resulting in poor consumer experiences and lost revenue. Additionally, the rise of online fraud through stolen passwords and device access complicates the protection of payment and device data.
Innovation Solution
Implementing automated fraud detection and analytics systems that aggregate transaction data from multiple payment vehicles and devices at the individual or household level, using a computing system to analyze transactions against fraud detection profiles, generate unique hash values, and notify financial institutions of potential fraudulent activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If transaction monitoring is performed at the individual card level, then the system is simple to implement, but the view of spending patterns is limited and fraudulent activities may be missed
Solution Approach 1:
The patent combines multiple card-level transaction datasets into a unified household-level profile, merging data from different payment vehicles and devices to create a comprehensive view of spending patterns. This aggregation enables more accurate fraud detection by analyzing transactions across all cards and devices associated with a household, rather than isolating each card individually.
Solution Approach 2:
The household-level fraud detection profile serves multiple functions: it monitors transactions across multiple cards, analyzes device usage patterns, detects fraudulent activities, and provides a unified view of household spending behavior. This multi-functional approach replaces the need for separate monitoring systems for each card, reducing overall system complexity while improving detection accuracy.
2Reliability
If suspicious transactions are declined without confirmation, then fraudulent activities are blocked, but consumer experience deteriorates and revenue opportunities are lost
Solution Approach 1:
The system uses household-level spending patterns as feedback to evaluate suspicious transactions. By comparing a transaction against the established household profile and spending behavior, the system can make more informed decisions about whether to approve or decline transactions, reducing false positives that harm consumer experience while maintaining fraud prevention effectiveness.
Solution Approach 2:
The system performs preliminary analysis of transactions against the household fraud detection profile before final approval or decline decisions. This preliminary action allows the system to identify obviously fraudulent transactions quickly while giving benefit of the doubt to transactions that align with household spending patterns, improving both fraud prevention and consumer experience.
3Measurement precision
If personal information is aggregated for fraud detection, then fraud detection accuracy improves, but user privacy concerns increase
Solution Approach 1:
The patent introduces an intermediary processing layer that aggregates transaction data and device information into household-level profiles without exposing raw personal information. This intermediary structure allows the system to analyze spending patterns and detect fraud while maintaining privacy by not requiring direct access to or storage of sensitive personal data, thus balancing detection accuracy with privacy protection.
4Reliability
If multiple payment vehicles and devices are monitored, then comprehensive fraud detection is achieved, but data processing complexity increases
Solution Approach 1:
The system segments data collection and processing into distinct components: transaction data from multiple cards, device information from various devices, and household profile data. By segmenting these data streams and processing them through standardized pipelines, the system can comprehensively monitor multiple payment vehicles and devices while managing complexity through modular architecture.
Data Source
AI summary
Systems and methods are disclosed for automated fraud detection and analytics using aggregated payment vehicles and devices, at the individual and/or household level. One method includes receiving an authorization request for a payment transaction originating at a merchant, using a first payment vehicle; receiving device information of a first device used in the payment transaction; retrieving transaction data and identifying information associated with the authorization request before the authorization request is routed to a financial institution; searching and determining payment vehicles and devices associated with the individual using the retrieved identifying information; aggregating transaction data associated with the payment vehicles and devices from the transaction database; retrieving reported fraudulent activities pertaining to the payment vehicles and devices; and generating a profile data for the individual according to the identifying information associated with the authorization request, personally identifiable information (PII), the aggregated transaction data, and reported fraudulent activities.