Efficient and Scalable Approach for Large Datasets Using Extended Version of K-Mediod Algorithm

Abstract

Amongst various kinds of clustering techniques, partition based clustering are of immense importance and they are widely used in the modern days large data analysis. The K-Means and K-Mediod Algorithms in their existing format carry certain weakness. For example in case of K-Means algorithm ‘deformation’ and deviations may arise due to the misbehaviour and disruption in the computing process. Similarly in case of K-Mediod Algorithm a lot of iteration is required which consumes huge amount of time and their by reduces the efficiency of clustering. In addition, scalability is also not suitably adapted in the data analysis. The present work seeks to modified and extend K-Mediod Algorithm to improve its efficiency in terms of reduction of execution time, so as to enable us to used it more effectively in case of Large datasets. The present study manifests that extended K-Mediod Algorithm are capable of handling large datasets with faster execution time, better scalability and overall improved efficiency.