Fast online anomaly detection using scan statistics
Ryan Turner, Steven Bottone and Zoubin Ghahramani
In: Machine Learning for Signal Processing (MLSP '10), 29 Aug 2010 - 1 Sep 2010, Kittilä, Finland.
We present methods to do fast online anomaly detection using scan
statistics. Scan statistics have long been used to detect statistically
significant bursts of events. We extend the scan statistics framework
to handle many practical issues that occur in application: dealing
with an unknown background rate of events, allowing for slow natural
changes in background frequency, the inverse problem of finding an
unusual lack of events, and setting the test parameters to maximize
power. We demonstrate its use on real and synthetic data sets with
comparison to other methods.