Single-Pass Graph Matching Using Alternating Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional matching systems for large graphs require substantial memory and computational resources, especially in streaming applications where data is constantly being produced, leading to inefficiencies and poor quality matches due to biased algorithms or the need for repeated full matching calculations.
Innovation Solution
A single-pass matching system that leverages an initial random subset of edges to generate an increased matching by selecting alternating paths, allowing for a more efficient and accurate matching of nodes in large graphs using limited memory, without compromising the quality of the matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional matching systems process large graphs with billions of nodes and edges, then matching accuracy can be maintained, but substantial storage space and computational time are required
Solution Approach 1:
The patent segments the large graph into smaller components by processing edges in streaming order and maintaining only the necessary matching information in memory. Instead of loading the entire graph, it processes edges sequentially and maintains matching states for individual nodes, dividing the computational problem into manageable units that fit within limited memory resources.
Solution Approach 2:
The patent performs preliminary actions by pre-processing edges to identify and store only the critical matching relationships. It uses a two-pass approach where the first pass identifies potential matches and the second pass refines them, performing necessary computations before the full matching is required to reduce subsequent processing demands.
2Measurement precision
If repeated full matching calculations are performed to improve match quality, then matching accuracy increases, but computational time and resources increase substantially
Solution Approach 1:
The patent applies partial action by performing matching calculations on a subset of edges and nodes at each step rather than repeatedly processing the entire graph. The streaming algorithm maintains incremental matching results, performing only the necessary local adjustments when new edges arrive, avoiding redundant full-graph computations while still achieving high-quality matches.
Solution Approach 2:
The patent ensures continuity of useful action by maintaining the matching state continuously as edges stream in, rather than performing discrete repeated full matching calculations. The algorithm continuously updates matching relationships incrementally, keeping the matching information current and accurate without the computational overhead of repeated complete recalculations.
3Productivity
If streaming algorithms use limited memory to process large data streams, then resource efficiency improves, but matching accuracy may be compromised
Solution Approach 1:
The patent extracts only the essential matching information from the streaming edges, storing minimal data structures that capture the critical relationships needed for accurate matching. It extracts and maintains node matching states and edge relationships in a compressed form, discarding redundant information that would consume memory without contributing to matching accuracy.
Solution Approach 2:
The patent changes parameters by using probabilistic data structures and approximation techniques that allow accurate matching with reduced memory footprints. It employs parameterized algorithms that can adjust the trade-off between memory usage and accuracy, using techniques like sampling and sketching to maintain matching quality while operating within limited memory constraints.
Data Source
AI summary
Embodiments of the present invention provide systems, methods, and computer storage media for determining an increased matching for large graphs in which an increased matching is generated for the graph by leveraging an initial matching for a small fraction of edges of the large graph. An initial matching for a random subset of edges of an input graph is leveraged to generate alternating paths based on the initially matched edges and the remaining edges, not included in the random subset. An increased matching for the entire graph includes the alternating paths without the initial matched edges, thus increasing the number of matched edges in the increased matching by at least one for every initially matched edge. Graph-based tasks may then be triggered based on the increased matching.