About Spatial Anomaly Detection

Anomaly detection identifies outliers and novelties, defined as observations that are significantly different from the others.

Also, note the following:

• Outlier detection estimators identify regions where the training data is concentrated, ignoring the deviant observations.
• Novelty detection identifies whether a new or unseen observation is an outlier according to an already defined training set.

Spatial anomaly detection identifies geographically isolated observations using spatial weights with standard anomaly detection methods. Examples include analyzing all environmental or traffic monitoring sensor data to find anomalies, which can lead to identifying dysfunctional sensors.