Joint Subgraph-to-Subgraph Transitions -- Generalizing Triadic Closure for Powerful and Interpretable Graph Modeling

Abstract

We generalize triadic closure, along with previous generalizations of triadic closure, under an intuitive umbrella generalization: the Subgraph-to-Subgraph Transition (SST). We present algorithms and code to model graph evolution in terms of collections of these SSTs. We then use the SST framework to create link prediction models for both static and temporal, directed and undirected graphs which produce highly interpretable results. Quantitatively, our models match out-of-the-box performance of state of the art graph neural network models, thereby validating the correctness and meaningfulness of our interpretable results.

arXiv commentPublished in WSDM 2021

References in corpus (6)

• Biological network comparison using graphlet degree distribution
• Motifs in Temporal Networks
• Evaluating Link Prediction Methods
• Triadic closure as a basic generating mechanism of communities in complex networks
• Triadic closure dynamics drives scaling-laws in social multiplex networks
• Directed closure measures for networks with reciprocity