Synthetic Merchant Authentication for Harder-to-Guess Account Access

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Authentication questions based on real or synthetic transactions are vulnerable to exploitation by malicious users due to the predictability and believability issues of synthetic transactions, particularly when they are not grounded in realistic merchant data.

Innovation Solution

Generate synthetic merchants by analyzing real merchant names using natural language processing techniques and machine learning models to create believable synthetic transactions, which are then used to formulate authentication questions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If synthetic transactions are used for authentication questions, then security against unauthorized access is improved, but the transactions become easily identifiable as fake reducing effectiveness

Engineering Contradiction:
Improveauthentication securityVSAvoiddetectability of synthetic transactions
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent creates synthetic merchants by copying and recombining name elements from real merchants. Instead of inventing completely fictional merchants, the system extracts meaningful components (prefixes, suffixes, location indicators) from authentic merchant names and assembles them into new, believable combinations that pass as real transactions

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system transforms real merchant name data by changing parameters such as combining different name elements, modifying location information, and adjusting transaction details to create synthetic versions that maintain statistical properties of real transactions while being distinguishable through analysis

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If real merchant information is used for authentication questions, then the questions are believable to users, but malicious users can exploit predictability to guess answers

Engineering Contradiction:
Improveuser recognition of real transactionsVSAvoidvulnerability to guessing attacks
Core Design Contradiction:
Ease of operationVSObject-affected harmful factors

Solution Approach 1:

The patent segments merchant names into distinct components (prefixes, suffixes, location elements) and recombines them to create synthetic merchants. This segmentation allows the system to maintain recognizable patterns from real merchants while creating unique combinations that don't correspond to actual transactions, preventing guessing attacks

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of using real merchant transactions directly or creating obviously fake merchants, the system inverts the approach by generating synthetic merchants that mimic real ones through systematic recombination of name elements, making them indistinguishable without analysis

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentEP4109307B1Account authentication using synthetic merchants
Publication Date: 2026.04.15 CAPITAL ONE SERVICES LLC
  • EP4109307B1 patent drawingFigure 1
  • EP4109307B1 patent drawingFigure 2
  • EP4109307B1 patent drawingFigure 3

AI summary

Methods, systems, and apparatuses are described herein for improving computer authentication processes through the generation of synthetic merchants. A plurality of different real merchant names may be received. The plurality of different real merchant names may be processed to determine one or more name elements. A request for access to an account associated with a user may be received. Based on the one or more name elements, one or more synthetic merchant names may be generated. Based on the one or more synthetic merchant names, synthetic transaction data may then be generated. A synthetic authentication question may be generated and presented to a user. A candidate response to the synthetic authentication question may be received. Based on the candidate response, access to the account may be provided.