The Random Plots Graph Generation Model for Studying Systems with Unknown Connection Structures

Area of Science:

• Graph theory
• Network science
• Complex systems modeling

Background:

• Modeling complex systems often lacks information on internal connection structures.
• Determining appropriate vertex degree distributions for graphs of such systems is challenging.

Purpose of the Study:

• Propose a novel method for modeling complex systems with unknown structures.
• Develop a random graph generator, Random Plots, to create diverse graph ensembles.

Main Methods:

• Generate a diversified set of vertex degree distributions.
• Employ a targeted graph generator for each distribution to create an ensemble.
• Formalize graph ensemble diversity using numerical characteristics.
• Compare Random Plots diversity against Erdos-Rényi-Gilbert (ERG), scale-free, and small-world models.

Main Results:

• Random Plots generates ensembles of graphs with diverse degree distributions.
• The proposed method demonstrates superior diversity compared to ERG, scale-free, and small-world models in most cases.
• Computational experiments validate the effectiveness of the Random Plots approach.

Conclusions:

• Random Plots offers a robust approach for generating diverse graph ensembles for complex systems.
• This method addresses the challenge of modeling systems with unknown connection structures.
• The generated ensembles provide a more representative range of possibilities for system analysis.