Separable models for dynamic signed networks

2025-05-12

AI Summary

This paper addresses the challenge of modeling relational polarity (support vs. opposition) in dynamic signed networks. We propose the first temporally separable Exponential Random Graph Model (ERGM) grounded in conditional independence assumptions, explicitly decoupling the generative mechanisms of polarity and interaction strength. Our method integrates a multilayer ERGM architecture, Bayesian inference, and an adaptive Metropolis–Hastings approximate exchange algorithm, enhancing flexibility for binary network modeling while preserving structural balance theory consistency. Applied to U.S. Senate roll-call voting data (1985–1989), our approach empirically identifies, for the first time within a dynamic framework, a stable bipartite structure: intra-party cohesion and inter-party antagonism—fully aligned with balance-theoretic predictions. The model delivers both statistical rigor and strong interpretability, offering a novel paradigm for analyzing the evolution of adversarial coalitions in complex systems.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Preferences
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Signed networks capture the polarity of relationships between nodes, providing valuable insights into complex systems where both supportive and antagonistic interactions play a critical role in shaping the network's dynamics. We propose a separable temporal generative framework based on multi-layer exponential random graph models, characterised by the assumption of conditional independence between the sign and interaction effects. This structure preserves the flexibly and explanatory power inherent in the binary network specification while adhering to consistent balance theory assumptions. Using a fully probabilistic Bayesian paradigm, we infer the doubly intractable posterior distribution of model parameters via an adaptive Metropolis-Hastings approximate exchange algorithm. We illustrate the interpretability of our model by analysing signed relations among U.S. Senators during Ronald Reagan's second term (1985-1989). Specifically, we aim to understand whether these relations are consistent and balanced or reflect patterns of supportive or antagonistic alliances.

Problem

• Modeling dynamic signed networks with separable temporal generative framework
• Inferring intractable posterior distribution using Bayesian methods
• Analyzing signed relations for consistency and balance in networks

Innovation

• Separable temporal generative framework for signed networks
• Multi-layer exponential random graph models
• Bayesian inference with adaptive Metropolis-Hastings algorithm