AdaGUM: An Adaptive Graph Updating Model-Based Anomaly Detection Method for Edge Computing Environment

Algorithm 2

Anomaly detection on edge nodes.

	Input: -the threshold of feature importance
	-the latest updating time
	-the time interval of updating
	-the collection of the data whose patterns are unknown
	-the feature graph of anomalous patterns
	-the feature graph of normal patterns
	Output: -the updated graph of anomalous patterns
	-the updated graph of normal patterns
(1)	;
(2)	;
(3)	;
(4)	;
(5)	while ((GetCurrentTime()-) )
(6)	for (each node in )
(7)	{if ()
(8)	{; }
(9)	for (each node in )
(10)	if ()
(11)	{; }
(12)	}
(13)	;
(14)	;
(15)	;
(16)	return;