Enhanced K Strange Points Clustering Algorithm
Terence Johnson, Santosh Kumar Singh · 2015
The algorithm proposed in this paper enhances the K Strange points clustering algorithm by selecting the first of unchanging K strange points as the minimum of the dataset and then finds the next strange point as the point which is farthest from the minimum and continues this process till it finds the K points which are farthest and almost equally spaced from each other. It then assigns the remaining points in the dataset into clusters formed by these K farthest or Strange points. The algorithm presented in this paper successfully addresses the issues related to longer execution time and formation of inaccurate clusters seen in the K Strange points clustering algorithm.