Predicting Metabolite-Disease Associations Based on Linear Neighborhood Similarity with Improved Bipartite Network Projection Algorithm

Algorithm 1

WKNKN.

	Input: matrixes , , , neighborhood sizes K, and decay term T.
	Output: new feature matrix MF.
	for
	mnn = KNN (p, , K)
	or to K do
	end for
	end for
	for to nd do
	dnn = KNN (q, , K)
	for to K do
	end for
	end for
	MF = max(M, )
	return MF
	end function