Latent Self-Exciting Point Process Model for Spatial-Temporal Networks
This addresses the challenge of incomplete interaction data in spatial-temporal networks, which is incremental as it extends existing point process models to handle latent participant information.
The paper tackles the problem of inferring unknown participants in spatial-temporal interaction events when only partial information is available, developing a latent self-exciting point process model that achieves promising results on identity-inference tasks and outperforms baselines in predicting future event timing and participants.
We propose a latent self-exciting point process model that describes geographically distributed interactions between pairs of entities. In contrast to most existing approaches that assume fully observable interactions, here we consider a scenario where certain interaction events lack information about participants. Instead, this information needs to be inferred from the available observations. We develop an efficient approximate algorithm based on variational expectation-maximization to infer unknown participants in an event given the location and the time of the event. We validate the model on synthetic as well as real-world data, and obtain very promising results on the identity-inference task. We also use our model to predict the timing and participants of future events, and demonstrate that it compares favorably with baseline approaches.