Modeling Social Networks with Node Attributes using the Multiplicative Attribute Graph Model

Abstract

Networks arising from social, technological and natural domains exhibit rich connectivity patterns and nodes in such networks are often labeled with attributes or features. We address the question of modeling the structure of networks where nodes have attribute information. We present a Multiplicative Attribute Graph (MAG) model that considers nodes with categorical attributes and models the probability of an edge as the product of individual attribute link formation affinities. We develop a scalable variational expectation maximization parameter estimation method. Experiments show that MAG model reliably captures network connectivity as well as provides insights into how different attributes shape the network structure.

15 pages, 7 figures, 7 tables

References in corpus (3)

• Kronecker Graphs: An Approach to Modeling Networks
• Multiplicative Attribute Graph Model of Real-World Networks
• Multifractal Network Generator

Cited by in corpus (12)

• NetGAN: Generating Graphs via Random Walks
• Multiplicative Attribute Graph Model of Real-World Networks
• An Infinite Latent Attribute Model for Network Data
• Detecting Strong Ties Using Network Motifs
• Nonparametric Multi-group Membership Model for Dynamic Networks
• Latent Space Model for Multi-Modal Social Data
• Bayesian Degree-Corrected Stochastic Blockmodels for Community Detection
• A Social Network Analysis of Articles on Social Network Analysis
• Quilting Stochastic Kronecker Product Graphs to Generate Multiplicative Attribute Graphs
• Node isolation in large homogeneous binary multiplicative attribute graph models
• Modeling Transitivity in Complex Networks
• Connectivity and giant component in random distance graphs