Knowledge Graph Versioning for Assumption Impact Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to provide information on how selection of assumptions influences conclusions in decision-making scenarios using knowledge graphs.
Innovation Solution
An information processing apparatus and method that generates and calculates scores for alternative knowledge graphs based on assumptions, using a link predicting technique to determine the influence of assumption selection on conclusion possibilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If knowledge graphs are used for decision-making assistance, then the ability to represent relationships between objects is improved, but the ability to provide information on how assumption selection influences conclusions is insufficient
Solution Approach 1:
The patent segments the knowledge graph into multiple versions by applying different assumptions to the same base graph. Each assumption creates a separate graph version, allowing independent analysis of how each assumption affects the conclusion. This segmentation enables the system to track and compare the influence of different assumptions without overwhelming complexity.
Solution Approach 2:
The patent performs preliminary actions by generating multiple knowledge graph versions with different assumptions before reaching the conclusion. By pre-computing the various graph configurations and their corresponding conclusion scores, the system prepares all necessary information in advance, making the assumption influence analysis efficient and comprehensive.
2Loss of information
If multiple knowledge graphs are generated for different assumptions, then the analysis of assumption influence is improved, but the computational complexity increases
Solution Approach 1:
The patent creates copies of the base knowledge graph for each assumption, generating multiple graph versions that can be processed in parallel. Instead of sequentially modifying the original graph, the system produces independent copies, each representing a different assumption scenario. This copying approach enables efficient computational distribution and reduces the burden on single-processing resources.
3Measurement precision
If link predicting techniques are applied to calculate conclusion scores, then the precision of conclusion evaluation is improved, but the processing time increases
Solution Approach 1:
The patent applies link predicting techniques selectively to specific links that are most influential in determining the conclusion, rather than computing scores for all possible links in the knowledge graph. By focusing computational resources on the most critical links, the system achieves sufficient precision for decision-making while reducing overall processing time.
Data Source
AI summary
To attain the object of generating information useful in a decision-making situation with use of a knowledge graph and providing the information to a user, at least one processor included in an information processing apparatus executes a graph editing process of generating at least one second knowledge graph by making editing corresponding to each of assumptions with respect to a first knowledge graph (for example, knowledge graph which deals with issues concerning medical care), a link predicting process of calculating, with use of a link predicting technique, a score of a link corresponding to a conclusion in each of the at least one second knowledge graph, and an information generating process of generating, with reference to the score, information which indicates an influence that selection of an assumption has on the conclusion.