Automatic Determination of Clustering Centers for “Clustering by Fast Search and Find of Density Peaks”

Algorithm 1

ACDPC.

	Input: datasets , parameter P;
Output: clustering result;
(1)	RhoSet = Ø, DeltaSet = Ø, and GammaSet = Ø;
//Part 1: Metric extraction
(2)	distanceMatrix = DistanceFunction (S); //Calculate distance according to equation (1);
(3)	Calculate the cutoff distance dc according to equation (3);
(4)	RhoSet = Fρ (distanceMatrix, dc); //Calculate ρ
(5)	DeltaSet = Fδ (distanceMatrix, Rhoset); //Calculate δ
(6)	GammaSet = Rhoset·DeltaSet; //
//Part 2: clustering center identification
(7)	= sort (GammaSet, “descend”); //Sort GammaSet in descending order to get a set of ordered statistics γ, indicates the subscript of GammaSet in descending order
(8)	Calculate DmTi according to equation (6);
(9)	Calculate the discrimination distance according to equation (8);
(10)	while i > 1 do
(11)	If ()
{
(12)	M = i;
}
(13)	Else
{
(14)	Break;
}
(15)	end
(16)	Identify the objects corresponding to {XT1, XT2,...., XT(i)} as the clustering centers {C1, C2,..., CM}, and label Ci as i;
//Part 3: Object clustering
(17)	for i = 1 to N do
(18)	if Xi is unlabeled then
(19)	Mark Xi the with label of its nearest neighbor with higher ρ;
(20)	end
(21)	end
(23)	return;