Dynamic Transaction Data Slicing for Authorization Rate Anomaly Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Electronic payment systems face challenges in identifying gradual changes in authorization rates due to anomalous behavior, which conventional data analytics methods struggle to detect efficiently, leading to delayed detection of system failures and revenue loss.

Innovation Solution

A computer model dynamically segments transaction data into slices using entropy and information gain values to identify anomalies, iteratively combining slices until the largest data slice exhibiting anomalous behavior is found, without requiring pre-configured attributes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is analyzed at a high level (aggregated data), then processing efficiency is improved, but anomaly detection capability deteriorates because anomalous behavior in small subsets is diluted

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidanomaly detection capability

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies segmentation by dividing the aggregated transaction data into multiple data slices based on different attributes (e.g., merchant category, transaction type, geographic location). This allows the system to maintain processing efficiency at the aggregated level while simultaneously enabling precise anomaly detection within specific subsets by analyzing each slice separately for deviations from expected patterns.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If data is segmented into many segments to improve anomaly detection, then anomaly identification precision is improved, but data sparsity increases and processing complexity increases

Engineering Contradiction:

Improveanomaly identification precisionVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamics by using a computer model that adaptively determines the optimal number and configuration of data slices based on the analysis results. The model dynamically adjusts the segmentation strategy, creating more slices when anomalies are detected in particular subsets and fewer slices when data appears homogeneous, thereby maintaining precision while managing processing complexity through adaptive rather than static segmentation.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent applies parameter changes by modifying the attributes and parameters used for data slicing based on feedback from anomaly detection. The computer model adjusts slicing parameters (such as which attributes to use for segmentation and what threshold criteria to apply) to optimize both anomaly detection precision and processing efficiency, avoiding fixed segmentation schemes that may create unnecessary complexity.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If conventional filtering and thresholding methods are used to identify drops, then processing simplicity is maintained, but detection capability for gradual shifts deteriorates

Engineering Contradiction:

Improveprocessing simplicityVSAvoidgradual shift detection capability

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical filtering and thresholding approach with a computer model-based statistical analysis system. This substitution enables the detection of gradual shifts through probabilistic modeling and pattern recognition, allowing the system to identify anomalies that result in smaller, more gradual drops in authorization rates while maintaining relative processing simplicity through automated model execution.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Speed

If static data slicing is used, then processing speed is maintained, but adaptability to different anomalous behaviors deteriorates

Engineering Contradiction:

Improveprocessing speedVSAvoidadaptability to anomalous behaviors

Core Design Contradiction:

SpeedVSAdaptability or versatility

Solution Approach 1:

The patent resolves this contradiction by implementing dynamic data slicing where the computer model adapts the slicing strategy based on the characteristics of anomalous behaviors detected. The system maintains processing speed through automated model execution while achieving high adaptability by adjusting slice configurations, attributes, and segmentation criteria in response to the specific patterns and types of anomalies identified in the transaction data.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS20250209058A1Methods and systems for authorization rate gradual degradation detection

Publication Date: 2025.06.26 STRIPE LLC

AI summary

Described herein are systems and methods to use modeling techniques to identify gradual changes in various metrics identified as a result of analyzing an aggregated transaction dataset. In one method, a computer model dynamically slice the data using an attribute, calculates an entropy value for using a rolling time window, and uses the entropy value to identify anomalous behavior. The model may use information gain to determine whether to further segmented the data slice into smaller data slices. The model may iteratively slice and analyze the data until a data slice corresponding to the root cause is determined. The model may then traverse the hierarchy of data slices and combine the data slices until an optimized combined data slice. The model may train a machine learning component, such as a booted tree algorithm, to optimize its traversal of the hierarchy of data slices.