A dimension reduction technique for K-Means clustering algorithm

To increase the efficiency of the clustering algorithms and for visualization purpose the dimension reduction techniques may be employed. In this paper our aim is to develop a simple dimension reduction technique to convert a high dimensional data to two dimensional data and then apply K-Means clustering algorithm on converted (two dimensional) data. We have applied our technique on three real datasets to evaluate the performance of our technique and for comparative purpose we have compared our technique with other existing technique.