Bayesian one-mode projection for dynamic bipartite graphs

TL;DR

This paper introduces a Bayesian approach for dynamic one-mode projection of bipartite graphs, capturing uncertainty and incorporating prior knowledge to improve robustness and scalability in evolving network analysis.

Contribution

It presents a novel Bayesian methodology with efficient update rules for dynamic bipartite graph projection, enhancing robustness to noise and scalability.

Findings

Provides probabilistic link presence estimates

Reduces sensitivity to noise and missing data

Scalable to large networks

Abstract

We propose a Bayesian methodology for one-mode projecting a bipartite network that is being observed across a series of discrete time steps. The resulting one mode network captures the uncertainty over the presence/absence of each link and provides a probability distribution over its possible weight values. Additionally, the incorporation of prior knowledge over previous states makes the resulting network less sensitive to noise and missing observations that usually take place during the data collection process. The methodology consists of computationally inexpensive update rules and is scalable to large problems, via an appropriate distributed implementation.