Temporal Neighbourhood Aggregation: Predicting Future Links in Temporal Graphs via Recurrent Variational Graph Convolutions

Bonner, S; Atapour, Abarghouei, A; Jackson, P; Brennan, J; Kureshi, I; Theodoropoulos, G; McGough, S; Obara, B

doi:10.1109/BigData47090.2019.9005545

Temporal Neighbourhood Aggregation: Predicting Future Links in Temporal Graphs via Recurrent Variational Graph Convolutions

Lookup NU author(s): Stephen Bonner ORCiD, Dr Amir Atapour Abarghouei ORCiD, Phillip Jackson, Dr John Brennan, Dr Stephen McGough, Professor Boguslaw Obara ORCiD

Downloads

Accepted version [.pdf]

Licence

This is the authors' accepted manuscript of a conference proceedings (inc. abstract) that has been published in its final definitive form by IEEE, 2019.

For re-use rights please refer to the publisher's terms and conditions.

Abstract

Graphs have become a crucial way to represent large, complex and often temporal datasets across a wide rangeof scientific disciplines. However, when graphs are used as input to machine learning models, this rich temporal information is frequently disregarded during the learning process, resulting in suboptimal performance on certain temporal inference tasks. To combat this, we introduce Temporal Neighbourhood Aggregation (TNA), a novel vertex representation model architecture designed to capture both topological and temporal information to directly predict future graph states. Our model exploits hierarchical recurrence at different depths within the graph to enable exploration of changes in temporal neighbourhoods, whilst requiring no additional features or labels to be present. The final vertex representations are created using variational sampling and are optimised to directly predict the next graph in the sequence. Our claims are supported by experimental evaluation on both real and synthetic benchmark datasets, where our approach demonstrates superior performance compared to competing methods, outperforming them at predicting new temporal edges by as much as 23% on real-world datasets, whilst also requiring fewer overall model parameters.