Transaction Graph Visualization for Fraud Detection Rule Strength

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional fraud detection systems often result in excessive false positives, leading to the rejection of authentic transactions and potential revenue loss, as they rely on analysts' experience to define policies without effectively distinguishing between fraudulent and authentic transactions.

Innovation Solution

The improved technique involves rendering transaction data as graphs to visually identify fraud detection rule strength, allowing users to select criteria that generate subgroups of transactions, thereby creating rules with lower false positive rates by distinguishing between fraudulent and authentic transactions through different graph styles and ratios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If fraud detection policies are defined based on analysts' experience to flag as many fraudulent transactions as possible, then the detection of fraudulent transactions is improved, but the false positive rate increases causing rejection of authentic transactions

Engineering Contradiction:

Improvefraud detection accuracyVSAvoidfalse positives

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent creates visual representations (graphs) of transaction data patterns that replicate and make tangible the abstract knowledge accumulated by analysts. By copying transaction characteristics into visual formats, the system enables analysts to intuitively identify fraud patterns without manually processing each transaction, thereby improving detection accuracy while reducing false positives through pattern recognition rather than rigid rule application

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical process of manual rule configuration with an automated visualization system. Instead of analysts manually setting thresholds and rules based on experience, the system automatically generates graphs from transaction data, substituting the mechanical rule-setting process with an automated visual analysis mechanism that reduces human error and bias in policy definition

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Object-affected harmful factors

If broad fraud detection rules are applied to catch more fraudulent requests, then fraudulent transaction processing is prevented, but legitimate transactions are also rejected causing revenue loss

Engineering Contradiction:

Improvefraudulent transaction processingVSAvoidrevenue loss

Core Design Contradiction:

Object-affected harmful factorsVSLoss of energy

Solution Approach 1:

The patent applies local quality by visualizing specific subsets of transaction data with distinct graphical representations. Different transaction types and patterns are rendered with different visual characteristics, allowing analysts to focus on specific local patterns (e.g., particular authentication factor combinations) that indicate fraud, rather than applying uniform broad rules to all transactions. This enables targeted detection that preserves legitimate revenue while blocking fraud

Inventive Principle:

Principle #3Local quality

3Reliability

If manual policy configuration is used to define fraud detection rules, then the system can be configured to reflect analyst expertise, but the process is time-consuming and subjective

Engineering Contradiction:

Improvepolicy configuration accuracyVSAvoidpolicy configuration time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by automatically generating visualizations of transaction data patterns before analysts need to configure policies. The system pre-processes transaction data into graphical representations that highlight potential fraud patterns, authentication factor distributions, and transaction characteristics. This preliminary visual preparation eliminates the need for analysts to manually explore raw data, significantly reducing configuration time while maintaining accuracy through automated pattern recognition

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10115108B1Rendering transaction data to identify fraud detection rule strength

Publication Date: 2018.10.30 EMC IP HLDG CO LLC

AI summary

Techniques of identifying fraud detection rule strength involve varying the rendering of a graph from transaction data. Along these lines, a rules server computer provides a general graph from a group of transaction entries defining a group of fraudulent and authentic transactions on an electronic display. A user defines selection criteria that the rules server computer applies to the group of transaction entries to generate a subgroup of transaction entries. From the subgroup of transaction entries, the rules server computer provides a focused graph on the electronic display from the subgroup of transaction entries defining a subgroup of the group of fraudulent and authentic transactions. A ratio of the number of fraudulent transactions to the number of authentic transactions represented in the focused graph identifies the strength of the selection criteria for use in a fraud detection rule.