To Glossary Lobby

Time Window Distribution

The Time Window Distribution model focuses on the timestamp and the occurrence of the events.

ADWIN 

The idea of ADWIN is to start from time window W and dynamically grow the window W when there is no apparent change in the context, and shrink it when a change is detected. The algorithm tries to find two subwindows of W – w_{0} and w_{1} that exhibit distinct averages. This means that the older portion of the window – w_{0} is based on a data distribution different than the actual one, and is therefore dropped.

time window distribution

Start Monitoring Your Models in Minutes

Book a demo