Dynamical Graph Echo State Networks with Snapshot Merging for Dissemination Process Classification
This work addresses the problem of classifying spreading patterns in temporal graphs for applications like information or disease modeling, representing an incremental improvement over prior reservoir computing approaches.
The authors tackled the Dissemination Process Classification (DPC) problem by proposing a model that combines snapshot merging data augmentation with Dynamical Graph Echo State Networks, achieving better classification performance on six benchmark datasets compared to existing methods.
The Dissemination Process Classification (DPC) is a popular application of temporal graph classification. The aim of DPC is to classify different spreading patterns of information or pestilence within a community represented by discrete-time temporal graphs. Recently, a reservoir computing-based model named Dynamical Graph Echo State Network (DynGESN) has been proposed for processing temporal graphs with relatively high effectiveness and low computational costs. In this study, we propose a novel model which combines a novel data augmentation strategy called snapshot merging with the DynGESN for dealing with DPC tasks. In our model, the snapshot merging strategy is designed for forming new snapshots by merging neighboring snapshots over time, and then multiple reservoir encoders are set for capturing spatiotemporal features from merged snapshots. After those, the logistic regression is adopted for decoding the sum-pooled embeddings into the classification results. Experimental results on six benchmark DPC datasets show that our proposed model has better classification performances than the DynGESN and several kernel-based models.