Recurrent Adaptive Classifier Ensemble for Handling Recurring Concept Drifts

Algorithm 2

RACE.

	Input: ( ) the streaming data chunks
	archive of ensemble models at time step t
	Diversity measure : Q Statistic
	Drift Detection Method Detect Drift
	Output:: the generalized ensemble model at each time step t
(1)	For each incoming data chunk do
(2)	Train new model with data chunk
(3)	Test with
(4)	drift  Detect Drift ( )
(5)	if drift == true
(6)	adapt models to current data
(7)	else
(8)	Update with to maximize diversity
(9)	End if
(10)	If | − 1|  t then
(11)	{}
(12)
(13)
(14)	Endif
(15)	Calculate diversity of models
(16)	Output
(17)	Endif