Stochastic Event Prediction via Temporal Motif Transitions

TL;DR

The paper introduces STEP, a novel framework for temporal link prediction that models event sequences as motif transitions governed by Poisson processes, improving accuracy and efficiency over existing methods.

Contribution

STEP reformulates temporal link prediction as a sequential forecasting problem using motif transitions, integrating motif-based features with graph neural networks.

Findings

Up to 21% average precision improvement over baselines.

Achieved 0.99 precision in next k sequential forecasts.

Demonstrated lower runtime than competing motif-aware methods.

Abstract

Networks of timestamped interactions arise across social, financial, and biological domains, where forecasting future events requires modeling both evolving topology and temporal ordering. Temporal link prediction methods typically frame the task as binary classification with negative sampling, discarding the sequential and correlated nature of real-world interactions. We introduce STEP (STochastic Event Predictor), a framework that reformulates temporal link prediction as a sequential forecasting problem in continuous time. STEP models event dynamics through discrete temporal motif transitions governed by Poisson processes, maintaining a set of open motif instances that evolve as new interactions arrive. At each step, the framework decides whether to initiate a new temporal motif or extend an existing one, selecting the most probable event via Bayesian scoring of temporal likelihoods and structural priors. STEP also produces compact, temporal motif-based feature vectors that can be concatenated with existing temporal graph neural network outputs, enriching their representations without architectural modifications. Experiments on five real-world datasets demonstrate up to 21% average precision gains over state-of-the-art baselines in classification and 0.99 precision in next $k$ sequential forecasting, with consistently lower runtime than competing motif-aware methods.