Graph Drawing with Resistance Distances for Improved Visualization

This paper introduces a novel approach to stress-based graph drawing using resistance distance, offering improvements over traditional shortest-path distance methods. The use of resistance distance, derived from the graph Laplacian, allows for a more accurate representation of global graph structure and enables efficient embedding in Euclidean space. The proposed algorithm, Omega, provides a scalable and efficient solution for network visualization, demonstrating better neighborhood preservation and cluster faithfulness. The paper's contribution lies in its connection between spectral graph theory and stress-based layouts, offering a practical and robust alternative to existing methods.