. Author manuscript; available in PMC: 2017 Nov 3.

Published in final edited form as: Data Min Knowl Discov. 2016 Feb 15;31(1):1–31. doi: 10.1007/s10618-016-0455-0

Table 3.

An algorithm to find iSuccess and dSuccess and compute S scores for all their exemplars

Procedure find_Scores (trainData)
Input: Labeled data, trainData;
Output: S scores for iSuccess and dSuccess, S_iSuccess and S_dSuccess;

1	for n ← 1 to size(trainData)
2	minD ← Nearest_Neighbor_Distance_D (trainData(n), trainData);
3	minI ← Nearest_Neighbor_Distance_I (trainData(n), trainData);
4	if (trainData(n).label == Nearest_Neighbor_D ().label &&
5	trainData(n).label != Nearest_Neighbor_I ().label )
6	S_dSuccess.add (minD / minI);
7	end if
8	if (trainData(n).label != Nearest_Neighbor_D ().label &&
9	trainData(n).label == Nearest_Neighbor_I ().label )
10	S_iSuccess.add (minD / minI);
11	end if
12	end for