Transaction Stream Classification for Credit Stacking Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Lenders face challenges in assessing credit stacking activities and accurately determining revenue streams of businesses, which can impact loan repayment capacity and recovery, due to the lack of effective tools for analyzing multiple lines of credit and loans simultaneously.

Innovation Solution

A system utilizing a machine learning component with a supervised neural network to analyze historical transaction data, applying transaction categorization and entity recognition models to identify revenue and non-revenue streams, and train a revenue stream classifier model to detect credit stacking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional manual assessment methods are used to evaluate business revenue streams and credit activities, then lenders can maintain simple assessment processes, but they cannot accurately detect credit stacking activities or assess true revenue streams

Engineering Contradiction:

Improvedetection accuracy of credit stackingVSAvoidcomplexity of analytics system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments transaction data into distinct categories (revenue streams, credit activities, loan payments) using machine learning classification. This segmentation enables precise detection of credit stacking by separating and analyzing different transaction types independently, resolving the contradiction between detection accuracy and system complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an analytics rules engine as an intermediary between raw transaction data and lender decision-making. This intermediary layer processes and interprets complex transaction patterns, providing accurate credit stacking detection while shielding lenders from the underlying system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If lenders analyze multiple lines of credit and loans simultaneously to assess credit stacking, then they can improve loan repayment assessment accuracy, but they face difficulty gathering accurate pictures of all revenue streams

Engineering Contradiction:

Improveaccuracy of loan repayment assessmentVSAvoidcompleteness of revenue stream information

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The analytics system performs multiple functions simultaneously: it identifies revenue streams, detects credit stacking activities, categorizes transactions, and assesses loan repayment capacity all within a single unified platform. This multi-functionality ensures complete information gathering while maintaining reliable assessment accuracy.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system continuously monitors transaction data and provides feedback on credit stacking detection and revenue stream identification. This feedback mechanism ensures that lenders receive updated, accurate information about borrower financial状况, improving both information completeness and assessment reliability.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If borrowers apply for multiple short-term business loans and lines of credit in short succession, then they can access needed capital for operations and expansion, but they negatively impact their ability to afford new payments and increase default risk

Engineering Contradiction:

Improveaccessibility of credit capitalVSAvoidaffordability of loan payments

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The analytics system performs preliminary analysis of borrower transaction history and credit patterns before new loan applications are processed. By detecting credit stacking activities in advance through pattern recognition, the system prevents borrowers from taking on unsustainable debt levels, thereby maintaining both credit accessibility and payment affordability.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12566978B2Analytics rules engine for select transaction identification

Publication Date: 2026.03.03 MASTERCARD INT INC

AI summary

A system is configured to retrieve a set of raw transaction data. A transaction categorization model is applied to the raw transaction data. The transaction categorization model infers a category from each transaction and labels each transaction with the inferred category. An entity recognition model is applied to the labelled transaction data. The entity recognition model extracts an entity from each transaction and labels each transaction with the extracted entity. The system generates a plurality of transaction streams from the labelled transactions based on the category and entity labels. The system also labels each transaction stream with either a revenue label or a non-revenue label based on an analysis of the types of transactions defining the transaction stream. The system trains a supervised-based neural network using the labelled transaction streams to generate a revenue stream classifier model.