Edge Bundling for Layered Graph Clutter Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graph visualization techniques, such as the Sugiyama algorithm, often produce aesthetically unappealing and confusing visualizations, especially for graphs with a large amount of detail, making it difficult for users to comprehend complex relationships and patterns.
Innovation Solution
A graph processing module that iteratively bundles edges, orders edge components, straightens edge bundles, separates edge components, and reroutes edges around node objects to improve the visual appearance and clarity of layered graphs, facilitating better understanding of graph structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the Sugiyama algorithm is used to visualize layered graphs, then the graph structure is organized in a hierarchical manner with reduced edge crossings, but the visualization becomes aesthetically unappealing and confusing when graphs have a large amount of detail
Solution Approach 1:
The patent applies edge bundling to merge multiple edges into unified visual bundles. Edges that would otherwise be drawn as separate lines are combined into bundled structures that maintain their hierarchical organization while reducing visual clutter. This merging operation directly addresses the technical contradiction by consolidating multiple edge representations into a more compact and aesthetically pleasing form, improving visual comprehension without sacrificing the underlying graph structure.
Solution Approach 2:
The patent segments edges into bundles based on their spatial and topological characteristics. By dividing the edge set into distinct bundles that can be processed and rendered separately, the system manages complexity more effectively. This segmentation allows the visualization to handle graphs with large amounts of detail by organizing edges into manageable groups that maintain structural information while reducing overall visual complexity.
2Device complexity
If automated techniques are used to organize the graph, then the graph structure is simplified, but patterns and relationships may be obscured by the organization process
Solution Approach 1:
The edge bundling technique merges edges while preserving their topological relationships and spatial patterns. The bundling process combines edges that share common characteristics without obscuring the underlying graph structure or patterns. By maintaining the hierarchical organization and edge connections within bundles, the system simplifies the visualization while preserving essential information about relationships and patterns in the graph data.
3Device complexity
If edges are bundled together to reduce clutter, then the visual appearance is improved, but edge crossings may increase
Solution Approach 1:
The patent implements edge bundling that combines multiple edges into unified structures while maintaining careful ordering to minimize crossings. The bundling operation merges edges that can be drawn together without creating excessive crossings, balancing visual simplicity with topological clarity. This selective merging approach reduces clutter in areas where edges can be bundled without significantly increasing crossings, while maintaining edge separation where crossings would be problematic.
Data Source
AI summary
A graph processing module is described for producing a visualization of a layered graph. The graph processing module operates by: bundling edges in the layered graph to produce respective edge bundles; ordering the edges associated with the edge bundles to reduce edge crossings; straightening the edges; and separating the edges in the edges bundles with respect to each other. The graph processing module can operate on a layered graph produced by the Sugiyama algorithm or some other technique, reducing clutter in the layered graph and thus improving a user's understanding of the layered graph.