Material Property Graph Search for Unknown Parameter Relationships
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search systems are limited in identifying unknown combinations of material property parameters with significant relationships due to the large number of arbitrary combinations, relying heavily on user knowledge and experience, and often miss relationships with weak relevance across technical fields.
Innovation Solution
A search system comprising a database, graph generator, and graph searcher that creates a graph of material property parameters with mutual relationships, allowing for the identification of unknown combinations based on known relationships, using directed and undirected graphs, conditional edges, and reliability information to enhance search accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If search is conducted across multiple technical fields to identify unknown combinations of material property parameters, then the coverage and completeness of discovered relationships is improved, but the number of arbitrary combinations to be evaluated increases dramatically
Solution Approach 1:
The patent segments the search process into two distinct phases: (1) a graph construction phase that organizes known relationships between material property parameters into a structured graph representation, and (2) a graph search phase that efficiently queries this pre-organized structure. This segmentation avoids the need to evaluate all possible combinations directly, instead leveraging the organized graph to quickly identify relevant relationships across multiple technical fields.
Solution Approach 2:
The patent performs preliminary action by pre-constructing a graph that encodes all known relationships between material property parameters before the actual search is executed. This pre-processing step organizes the knowledge base in advance, allowing subsequent searches to efficiently traverse only relevant paths rather than evaluating all possible combinations from scratch.
2Adaptability or versatility
If the number of material property parameters is increased to cover multiple technical fields, then the comprehensiveness of the search is improved, but the computational burden and difficulty of selection increase
Solution Approach 1:
The patent introduces a graph structure as an intermediary representation between the raw material property parameters and the search query. This graph acts as a mediator that pre-processes and organizes the relationships among numerous parameters, transforming the complex selection problem into a simpler graph traversal problem. The graph structure makes the relationships among parameters visually and computationally manageable.
Solution Approach 2:
The patent replaces the mechanical approach of systematically evaluating all parameter combinations with a graph-based information retrieval approach. Instead of brute-force computation, the system uses graph algorithms to efficiently navigate and detect relationships, substituting computational mechanics with information-theoretic methods.
3Measurement precision
If users are expected to possess extensive knowledge and experience to select promising combinations, then the accuracy of combination selection is improved, but the accessibility and ease of use deteriorates
Solution Approach 1:
The patent implements self-service by enabling the system to automatically perform the knowledge-intensive task of identifying promising parameter combinations through graph search algorithms. Users do not need to manually apply their expertise to select combinations; instead, the system autonomously traverses the graph to detect relationships, making advanced material property analysis accessible to users regardless of their domain expertise.
Data Source
AI summary
Provided are a search system and a search method capable of searching an unknown combination of material property parameters having a significant relationship based on already known relationships from among multiple arbitrary combinations of material property parameters. The search system includes a database, a graph generator, and a graph searcher, and is configured as follows: the database stores the plurality of pairs of material property parameters having mutual relationships, and the graph generator generates a graph in which material property parameters stored in the database are nodes, and an inter-node corresponding to the pairs of material property parameters stored to have mutual relationships is an edge. The graph searcher searches the graph generated by the graph generator based on a given search condition and outputs a search result.