Multiple Network Embedding for Anomaly Detection in Time Series of Graphs

This paper considers the graph signal processing problem of anomaly detection in time series of graphs. We examine two related, complementary inference tasks: the detection of anomalous graphs within a time series, and the detection of temporally anomalous vertices. We approach these tasks via the adaptation of statistically principled methods for joint graph inference, specifically multiple adjacency spectral embedding (MASE) and omnibus embedding (OMNI). We demonstrate that these two methods are effective for our inference tasks. Moreover, we assess the performance of these methods in terms of the underlying nature of detectable anomalies. Our results delineate the relative strengths and limitations of these procedures, and provide insight into their use. Applied to a large-scale commercial search engine time series of graphs, our approaches demonstrate their applicability and identify the anomalous vertices beyond just large degree change.