Detecting Insider Threat from Behavioral Logs Based on Ensemble and Self-Supervised Learning

Algorithm 1

Training process for ensemble detector with Over-Bootstrap strategy.

	input: Set to legitimate sessions ; Set of legitimate sessions ; T: Number of subdetectors; iter:Iterations number; m:Batch size
Output: Ensemble detectors
(1)	: Initialize:fordo rand_init
(2)	:  fordo
(3)	:  fordo
(4)   Over-Bootstrap sampling:
(5)
(6)
(7)
(8)   Train for a single subdetector
(9)   Train
(10)  end for
(11) end for