Quantum-Based Feature Selection for Multiclassification Problem in Complex Systems with Edge Computing

Algorithm 2

Quantum ReliefF algorithm.

(1)	Init WT = (0, …, 0)^T
(2)	Normalized the sample sets:
(3)	Prepare quantum states for all samples by operations CMP and Ry, respectively.
(4)	for t = 1 to T do
(5)	Select a state |ϕ〉 from randomly which corresponds to u
(6)	Perform swap operation on |ϕ〉 and obtain
(7)	The similarity information coded into quantum state through swap test, the inner product and amplitude estimation operations
(8)	The nearest k samples in each class are obtained by Grover-Long method
(9)	for i = 1 to N do
(10)
(11)	end
(12)	end
(13)
(14)	for i = 1 to N do
(15)	if then
(16)	The i-th feature is relevant
(17)	else
(18)	The i-th feature is not relevant
(19)	end
(20)	end