Two classes of bipartite networks: nested biological and social systems

TL;DR

This paper introduces a model that classifies bipartite networks into biological and social classes based on their topology, and demonstrates its effectiveness in fitting real-world mutualistic and social network data.

Contribution

The study presents a generalized model for bipartite network evolution that distinguishes biological and social systems, providing both quantitative and qualitative insights.

Findings

Model accurately fits biological mutualistic network data

Model explains differences in social and biological bipartite networks

Statistical properties of projected graphs are analyzed

Abstract

Bipartite graphs have received some attention in the study of social networks and of biological mutualistic systems. A generalization of a previous model is presented, that evolves the topology of the graph in order to optimally account for a given Contact Preference Rule between the two guilds of the network. As a result, social and biological graphs are classified as belonging to two clearly different classes. Projected graphs, linking the agents of only one guild, are obtained from the original bipartite graph. The corresponding evolution of its statistical properties is also studied. An example of a biological mutualistic network is analyzed in great detail, and it is found that the model provides a very good quantitative fitting of its properties. The model also provides a proper qualitative description of the statistical features observed in social webs, suggesting the possible reasons underlying the difference in the organization of these two kinds of bipartite networks.