Physical-Digital Environment Associations Using Semantic Graph Tuples

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The human inverse base-rate effect in mixed reality environments leads to errors in rendering accurate information, as humans tend to pick rare outcomes when combining learned physical and virtual environments, causing a disconnect between the two.

Innovation Solution

Generating associations between physical and digital environments using relationship feature-sets comprising subjects, predicates, and objects, forming tuples to create strong associations and mitigate the inverse base-rate effect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple environments (physical and digital) are combined to provide rich information experiences, then the versatility and information richness are improved, but the accuracy of information rendering deteriorates due to human inverse base-rate effect

Engineering Contradiction:

Improveenvironment combination capabilityVSAvoidinformation rendering accuracy

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces relationship feature-sets as an intermediary mechanism that mediates between physical and digital environments. These feature-sets include relationship tuples (subject, predicate, object) that establish structured connections between entities across different environments, enabling accurate information rendering by explicitly defining relationships rather than relying on human perception alone

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameters of environment representation by introducing structured relationship feature-sets with specific tuples. This transforms the representation from unstructured visual data to structured semantic data, allowing the system to overcome human cognitive biases through computational relationship modeling

Inventive Principle:

Principle #35Parameter changes

2Reliability

If relationship feature-sets with multiple tuples are generated to improve association accuracy, then the information rendering accuracy is improved, but the computational complexity increases

Engineering Contradiction:

Improveassociation accuracyVSAvoidcomputational complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the complex task of environment association into discrete relationship feature-sets, each containing specific tuples (subject, predicate, object). This segmentation allows the system to process and manage associations in manageable units rather than handling the entire complex relationship space at once, reducing computational complexity while maintaining accuracy

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20250316080A1Generation of associations between physical and virtual environments

Publication Date: 2025.10.09 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments related to generation of associations between a physical environment and a digital environment are provided. A relationship feature-set is extracted for each of a first scene of the physical environment and a second scene of the digital environment. The relationship feature-set includes a subject, a predicate and an object, for each of the first scene and the second scene. The first scene is represented by a first tuple and the second scene is represented by a second tuple. The first tuple and the second tuple are used to generate a graph network, which is used to determine an adjacency of the first tuple and the second tuple in the graph network. The adjacency is used to calculate an association between the first scene and the second scene. An output is generated based on the association.