Scalable Hypergraph Visualization

Area of Science:

• Computer Science
• Data Visualization
• Network Analysis

Background:

• Hypergraph visualization is crucial for network data analysis.
• Polygon-based hypergraph representations offer benefits but face challenges with large datasets due to self-intersections.
• Existing methods struggle with scalability and layout optimization for complex hypergraphs.

Purpose of the Study:

• To propose a novel framework for improving hypergraph visualization.
• To address the issue of excessive self-intersections in polygon-based hypergraph layouts for large datasets.
• To develop an iterative simplification and layout optimization approach for enhanced hypergraph representation.

Main Methods:

• Iterative simplification of hypergraphs using atomic operations.
• Optimization of the layout for the simplified hypergraph.
• Reverse process to reconstruct the original hypergraph with an improved layout.
• Introduction of hypergraph planarity definitions and conditions within the polygon representation.

Main Results:

• A framework for iterative hypergraph simplification and layout optimization.
• A method to guide simplification using an operation priority measure.
• Demonstrated utility of the approach on real-world application datasets.
• Extension to handle simultaneous simplification and layout optimization for hypergraphs and their duals.

Conclusions:

• The proposed framework effectively reduces self-intersections in polygon-based hypergraph layouts.
• The iterative simplification and optimization method enhances the clarity and scalability of hypergraph visualization.
• The approach provides a robust solution for analyzing complex network data.