Runtime Enforcement of Partnership Conditions From Natural Language

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Customizing software code to manage and validate computer-detected events based on specific conditions is time-consuming, labor-intensive, and error-prone, leading to scalability and throughput issues.

Innovation Solution

A system extracts conditions from natural language documents using machine learning, generates machine-readable data structures, and validates events in real-time to reduce storage and network resource demands.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If software code is customized to manage and validate computer-detected events based on specific conditions, then the system can accurately enforce partnership agreement terms, but the implementation becomes time-consuming, labor-intensive, and error-prone

Engineering Contradiction:
Improveaccuracy of condition enforcementVSAvoidimplementation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent creates machine-readable copies of partnership agreement conditions from natural language documents. These structured data representations can be directly processed and enforced by software systems without requiring manual code customization, thus maintaining enforcement accuracy while dramatically reducing implementation time and human effort.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces an intermediary processing layer that converts natural language conditions into structured machine-readable formats. This intermediary step acts as a mediator between the human-written agreements and the software enforcement mechanisms, eliminating the need for direct code customization while ensuring accurate condition representation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If software code is customized to capture various attributes specific to partnership conditions, then the system can precisely validate events, but the process becomes vulnerable to errors and difficult to maintain

Engineering Contradiction:
Improveprecision of event validationVSAvoidsoftware complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the representation of conditions by changing parameters from unstructured natural language to structured machine-readable formats with defined schemas. This parameter transformation maintains validation precision while reducing software complexity, as the structured formats can be processed through standardized validation logic rather than custom-coded solutions.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments partnership agreement conditions into discrete, structured data elements that can be independently processed and validated. This segmentation allows the system to handle complex validation requirements through modular, standardized operations rather than monolithic custom code, reducing both complexity and error susceptibility.

Inventive Principle:
Principle #1Segmentation

3Loss of information

If traditional methods are used to manage partnership conditions, then detailed tracking is possible, but storage needs and network traffic increase, reducing scalability

Engineering Contradiction:
Improvecompliance tracking accuracyVSAvoidstorage resources
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent changes the data representation parameters from detailed natural language storage to compact structured formats. This transformation maintains all necessary compliance information while significantly reducing storage requirements and network transmission overhead, enabling scalable deployment across distributed systems.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20260064966A1Partner management runtime enforcement
Publication Date: 2026.03.05 CITIGROUP
  • US20260064966A1 patent drawing
  • US20260064966A1 patent drawing
  • US20260064966A1 patent drawing

AI summary

Context information applicable to an operation is obtained. A first computer-readable data structure encoding a first set of conditions applicable to performance of the operation is generated based on the context information. A second computer-readable data structure corresponding to an identifier of an entity for which the operation is performable is obtained, where the second computer-readable data structure encodes a second set of conditions applicable to the performance of the operation. A third computer-readable data structure is generated based on a third set of conditions encoded by both the first computer-readable data structure and the second computer-readable data structure. The third computer-readable data structure is caused to be used to apply the third set of conditions to control the performance of the operation for the entity.