A flow chart of the new GraphOptima framework's main modules: the Optimizer to explore the objective space of various solutions; the Layout Generator(s) to create a network layout based on a set of parameters; and the Layout Evaluator(s) to assess the readability of the resulting layout based on a set of readability metrics.
A scatter plot depicting the "Objective Space of Layout Solutions for a Randomly Generated ‘Price’ Network"
The graphic shows the Pareto front as a scatter plot, representing possible layout solutions for a sample network with 10,000 nodes and 9,999 edges. The network is generated using a preferential attachment model. Each dot represents a ForceAtlas2 layout with specific parameter values for Scaling Factor, Gravity, and Max Iter (the number of iterations). The X, Y, and Z coordinates correspond to the values of the three readability metrics used for optimization: Crosslessness, Normalized Edge Length Variance, and Minimum Angle.
🚨[New Year, New Paper] 🚨Introducing "GraphOptima: A graph layout optimization framework for visualizing large networks (500k+ nodes/ 1M+ edges)" #SNA #SocialNetworkAnalysis #NetworkScience #GraphViz #HighPerformanceComputing www.sciencedirect.com/science/arti...