Correlithm Object Processing for Non-Exact Data Similarity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional computers are limited in comparing and determining similarity between data samples due to their reliance on ordinal numbers, which only provide information about sequence order, making it difficult to identify similarities or matches without exact matches, especially in applications like face recognition and fraud detection, leading to complex processes that consume processing power and reduce speed and performance.

Innovation Solution

The implementation of a correlithm object processing system that uses categorical numbers and correlithm objects to represent data samples, enabling non-binary comparisons and quantifying similarity between data samples, allowing for direct comparison regardless of data type or format through the use of sensor, node, and actor tables.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional computers use ordinal numbers to represent data samples, then they can perform basic operations like counting and sorting, but they cannot determine similarity between data samples without exact matches

Engineering Contradiction:

Improvesimilarity detection capabilityVSAvoidsignal processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the numerical representation system from ordinal to categorical. Instead of using ordinal numbers where only sequence matters, the system uses categorical numbers that can represent similarity relationships. This parameter change in the fundamental data representation enables direct similarity comparison without complex signal processing.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces complex mechanical signal processing techniques with a simpler categorical number system. Instead of using traditional signal processing algorithms to determine similarity, the system uses categorical numbers that inherently encode similarity information, allowing direct comparison through simple equality checks.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If conventional computers rely on complex signal processing techniques to determine similarity, then they can identify matching data samples, but processing speed and performance are reduced

Engineering Contradiction:

Improvedata sample comparison accuracyVSAvoidprocessing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent performs preliminary encoding of similarity information during data representation. By embedding similarity relationships directly into the categorical number structure before comparison operations, the system eliminates the need for complex runtime signal processing, thereby improving processing speed while maintaining comparison accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates categorical number representations that copy and encode similarity relationships directly into the data structure. Instead of computing similarity through complex processing, the system uses pre-encoded categorical representations that can be compared directly, significantly improving processing speed.

Inventive Principle:

Principle #26Copying

3Adaptability or versatility

If conventional computers use ordinal binary integers for all operations, then they can maintain numerical order for calculations, but they lose the ability to represent similarity relationships between data samples

Engineering Contradiction:

Improvedata representation flexibilityVSAvoidsimilarity information

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent changes the fundamental parameter of numerical representation from ordinal to categorical. This allows the system to represent both numerical values and similarity relationships simultaneously, preventing loss of similarity information while maintaining computational capability.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent creates a universal categorical number system that can represent multiple types of information including numerical values and similarity relationships. This multi-functional representation system eliminates the need for separate encoding schemes and prevents information loss.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10789081B2Computer architecture for emulating drift-between string correlithm objects in a correlithm object processing system

Publication Date: 2020.09.29 BANK OF AMERICA CORP

AI summary

A device that includes a node engine configured to define a number of child correlithm objects for a string correlithm object. The node engine is further configured to set a starting correlithm object as a first parent correlithm object and set an ending correlithm object as a second parent correlithm object. The node engine is further configured to randomly select a correlithm object less than the standard distance away from the first parent correlithm object, define the selected correlithm object as a child correlithm object, and link the child correlithm objects with the first parent correlithm object. The node engine is further configured to randomly select a correlithm object less than the standard distance away from the second parent correlithm object, define the selected correlithm object as a child correlithm object, and link the child correlithm objects with the second parent correlithm object.