A Novel Three-Way Decision Model for Improving Computational Thinking Based on Grey Correlation Analysis

Algorithm 1

Calculation of weights for expert clustering.

	Input: Initial scoring of metrics by experts , n
	Output: Weight of metrics
(1)	calculate expert similarity matrix
(2)
(3)	for i = 1 to n do
(4)	for j = 2 to n do
(5)
(6)	cluster similar experts
(7)	// Initialize the largest collection and class among experts.
(8)	// Initialize the maximum number of similar classes among experts.
(9)	if find and then
(10)	corresponding two experts
(11)
(12)
(13)	Repeat the above steps until .
(14)	// Initialize the expert collection.
(15)	for i = 1 to k-1 do
(16)	for j = 2 to k do
(17)	combine the collections containing the same experts in the pairwise clusters
(18)	if unclustered experts then
(19)	separate into a class
(20)	// the weight of each class
(21)	// the metric evaluation vector of expert
(22)	// the entropy weight of the th expert in the class
(23)	// expert aggregate weights
(24)
(25)
(26)	// metric weights