Ensemble Merchant Prediction System for Transaction Data Grouping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing merchant prediction systems in the payment card industry face challenges in automatically determining group memberships for merchants based on transaction data, often requiring manual inspection and lacking explicit definitions for relationships between merchants and issuers.

Innovation Solution

An ensemble merchant prediction system utilizing multiple algorithms to analyze transaction data, assign confidence scores, and aggregate predictions to determine group memberships, including k-similar locations, aggregated locations as documents, third-party data prediction, numerical signature prediction, and statistical modeling to classify merchant locations and assign confidence values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual inspection of transaction data is used to discover merchant group relationships, then the accuracy of group membership determination can be maintained through expert judgment, but the productivity and automation extent are severely reduced

Engineering Contradiction:

Improveaccuracy of group membership determinationVSAvoidefficiency of merchant grouping

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual inspection (mechanical human analysis) with an automated prediction system that uses multiple algorithms including k-similar locations, aggregated locations as documents, third-party data prediction, numerical signature prediction, and statistical modeling. This substitution maintains accuracy while dramatically improving productivity by automating the merchant grouping process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the prediction algorithms to automatically analyze transaction data and determine merchant group memberships without requiring manual expert intervention. The ensemble prediction system autonomously processes data, generates predictions, and assigns confidence values, making the entire merchant grouping process self-sufficient.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If multiple prediction algorithms are used to improve the accuracy of merchant group predictions, then the measurement precision increases, but the device complexity increases due to the ensemble system requirements

Engineering Contradiction:

Improveaccuracy of merchant group predictionsVSAvoidcomplexity of prediction system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple independent prediction algorithms into a unified ensemble system that processes transaction data collectively. By combining k-similar locations, aggregated locations as documents, third-party data prediction, numerical signature prediction, and statistical modeling algorithms, the system achieves higher prediction accuracy while managing complexity through integrated processing.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The ensemble prediction system serves multiple functions simultaneously: it performs k-similar location analysis, aggregated location documentation, third-party data integration, numerical signature prediction, and statistical modeling all within a single unified framework. This multi-functionality improves prediction accuracy without proportionally increasing system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If confidence values are assigned to each predicted group membership to improve reliability, then the measurement precision and reliability increase, but the loss of time increases due to additional processing requirements

Engineering Contradiction:

Improveconfidence in group membership predictionsVSAvoidprocessing time for prediction

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-processing transaction data and pre-computing features such as k-similar locations, aggregated locations, and numerical signatures before making predictions. This preliminary preparation enables faster confidence value assignment during the actual prediction phase, reducing the time loss while maintaining high reliability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The ensemble prediction system maintains continuous useful action by processing multiple algorithms in parallel and continuously generating predictions with confidence values. The system avoids idle time by ensuring that all prediction components work continuously on the transaction data, efficiently producing reliable results without unnecessary delays.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS8738486B2Methods and apparatus for implementing an ensemble merchant prediction system

Publication Date: 2014.05.27 MASTERCARD INT INC

AI summary

A computer-based method for discovering patterns in financial transaction card transaction data for the purpose of determining group membership of a merchant within the transaction data is described. The data relates to merchants that accept the financial transaction card for payment. The method includes receiving transaction data from at least one database, predicting a membership of a merchant in a group using at least one prediction algorithm and the retrieved transaction data, the algorithm generating meta-data describing the predictions, inputting the at least one predicted group membership and the meta-data into a data mining application, and assigning a confidence value to each predicted group membership by the data mining application, utilizing the predicted group memberships and the meta-data.