Partition Selection for Large-Scale Data Management Using KNN Join Processing

Algorithm 3

Relevant partition selecting.

	Input: s,
	Output:
	selectTrainSet() {
	; //determine the initial Relevant-partition
	for i = 2 to k do
	; //calculate the k nearest neighbors of s in
	; //calculate the maximum distance from s to its k nearest neighbors from R
	; //calculate the distance from s to ,
	if () then// does not belong to
	break;
	else// belongs to
	; //add the partition to
	}