A Generative Model for Dynamic Networks with Applications

TL;DR

This paper introduces a neural network-based generative model for dynamic networks with evolving community structures, enabling improved community detection and link prediction in temporally changing social and collaboration networks.

Contribution

It presents a novel latent space model for dynamic networks with variable community counts over time, using neural networks for approximate inference.

Findings

Effective in community detection on synthetic and real networks

Outperforms existing approaches in link prediction tasks

Demonstrates utility in modeling temporal network dynamics

Abstract

Networks observed in real world like social networks, collaboration networks etc., exhibit temporal dynamics, i.e. nodes and edges appear and/or disappear over time. In this paper, we propose a generative, latent space based, statistical model for such networks (called dynamic networks). We consider the case where the number of nodes is fixed, but the presence of edges can vary over time. Our model allows the number of communities in the network to be different at different time steps. We use a neural network based methodology to perform approximate inference in the proposed model and its simplified version. Experiments done on synthetic and real world networks for the task of community detection and link prediction demonstrate the utility and effectiveness of our model as compared to other similar existing approaches.