Knowledgeable Machines Ontological Subject Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently identifying and evaluating the value significance and relationships within vast, unstructured bodies of knowledge, making it difficult to accelerate research, knowledge discovery, and decision-making processes.
Innovation Solution
The development of methods and systems that transform textual compositions into Participation Matrices and Association Strength Matrices, allowing for the evaluation of Value Significance Measures (VSMs) of ontological subjects, which in turn facilitates the identification of important concepts and their relationships within a system of knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional methods are used to process vast bodies of knowledge, then human processing capability is limited by memory capacity and processing speed, but the accuracy and speed of research and knowledge discovery can be improved by using automated systems
Solution Approach 1:
The patent introduces an automated knowledge processing system that acts as an intermediary between vast bodies of unstructured knowledge and human researchers. The system includes modules for identifying ontological subjects, constructing knowledge graphs, and evaluating value significances, thereby mediating the information processing task to overcome human cognitive limitations while preserving useful information.
Solution Approach 2:
The patent replaces the mechanical human brain processing system with an automated computational system. The automated system uses algorithms to identify ontological subjects, build knowledge graphs, and evaluate information significance, substituting human memory and processing capabilities with machine-based automated analysis to improve speed and accuracy.
2Measurement precision
If all parts of a system are analyzed in detail, then the accuracy of knowledge representation improves, but the complexity of processing increases significantly
Solution Approach 1:
The patent segments the complex task of knowledge analysis into distinct modules: identification of ontological subjects, construction of knowledge graphs, and evaluation of value significances. Each module handles a specific aspect of the analysis, allowing detailed processing of individual components while managing overall system complexity through modular architecture.
Solution Approach 2:
The patent applies local quality by evaluating the value significance of different ontological subjects and knowledge components differently. Rather than treating all parts uniformly, the system identifies and prioritizes significant portions of knowledge based on their importance to the research domain, enabling accurate representation of critical information while reducing processing burden on less significant elements.
3Loss of information
If unstructured textual compositions are processed directly, then the original information is preserved, but the clarity and usefulness of the knowledge for non-experts decreases
Solution Approach 1:
The patent introduces a knowledge graph construction module that acts as an intermediary representation layer between unstructured textual compositions and end users. The knowledge graph organizes information from text into structured visual representations showing relationships between ontological subjects, preserving the original information while making it accessible and understandable for non-experts through intuitive graphical interfaces.
Data Source
AI summary
The present invention discloses methods, systems, and tools to extract the usable knowledge from a body of knowledge and build computer/machine useable data structures stored in one or more non-transitory storage media. The body of knowledge is regarded as a composition of ontological subjects (OSs) of different orders. Using the participation information of the OSs into each other, one or more association strength matrices and/or conditional occurrence probability matrices and/or ontological subject maps are built from which the value significance, information content, and type and strength of the relationship of the partitions (i.e. OSs of different orders) of the composition are calculated and learned. The methods systematically build one or more data structures carrying the actionable knowledge from a body of knowledge and enables one to build knowledgeable and context aware systems and machines for various desired applications. Exemplary systems and machines, for implementing the methods and some exemplary applications and services, are disclosed.