Anomaly Detection and Localisation using Mixed Graphical Models
This work addresses anomaly detection for data streams with mixed variable types, but it is incremental as it builds on existing graphical models and CUSUM algorithms.
The paper tackles anomaly detection and localization in heterogeneous data streams using a mixed graphical model, achieving more precise variable-level anomaly identification compared to univariate marginal methods.
We propose a method that performs anomaly detection and localisation within heterogeneous data using a pairwise undirected mixed graphical model. The data are a mixture of categorical and quantitative variables, and the model is learned over a dataset that is supposed not to contain any anomaly. We then use the model over temporal data, potentially a data stream, using a version of the two-sided CUSUM algorithm. The proposed decision statistic is based on a conditional likelihood ratio computed for each variable given the others. Our results show that this function allows to detect anomalies variable by variable, and thus to localise the variables involved in the anomalies more precisely than univariate methods based on simple marginals.