Segment Matching Search Using MatchDiagrams for Large Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search technologies face challenges in efficiently searching large data sets due to computational inefficiencies and the inability to scale with graph-based representations, particularly in handling datasets exceeding main memory capacity, and the need for specialized hardware or software.
Innovation Solution
A segment matching search system that encodes graph search in a decision diagram, utilizing a MatchDiagram data structure, which is a directed acyclic graph representing matches between search patterns and input elements, allowing for efficient management of search results and optimization of search plans without requiring specialized hardware or software.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If graph-based search processes are used, then search query expressiveness is improved, but computational efficiency deteriorates due to NP-complete complexity
Solution Approach 1:
The patent segments the graph search process into distinct phases: pattern compilation into search plans, and execution of those plans. By dividing the complex graph search into manageable segments with predefined operations, the system maintains query expressiveness while improving computational efficiency through structured processing.
Solution Approach 2:
The patent performs preliminary actions by compiling search patterns into optimized search plans before execution. This pre-processing step transforms complex query expressions into efficient execution sequences, allowing the system to handle expressive queries without suffering from NP-complete complexity during actual search execution.
2Speed
If complete graph data structures are maintained in main memory, then search speed is improved, but device complexity increases and large datasets cannot be handled
Solution Approach 1:
The patent extracts only the necessary portions of graph data into main memory as MatchDiagrams during search execution, rather than maintaining complete graph data structures. This extraction approach enables fast in-memory processing while avoiding the complexity of managing large-scale graph data in memory, allowing datasets to exceed main memory capacity.
Solution Approach 2:
The patent introduces MatchDiagrams as intermediary data structures that bridge between the full graph data stored externally and the processing requirements of graph search. These intermediaries contain only the relevant match information needed for current searches, reducing memory complexity while maintaining search speed.
3Productivity
If specialized hardware or software is used, then handling of large searches is improved, but ease of operation deteriorates
Solution Approach 1:
The patent creates a universal search system that can handle various types of graph searches and large datasets through a standardized interface. The MatchDiagram framework and search plan compilation approach provide multi-functional capabilities that work across different graph types and sizes, eliminating the need for specialized hardware or software while maintaining ease of operation.
Data Source
AI summary
A planning and search system are described wherein a graph search and segment matching are used to handle very large searches at a higher speed.