Ayush Tiwari
Clustering plays a very important role in research area in the field of data mining. Clustering is a method of subdivide a set of data in an important sub classes called clusters. It assists users to recognize the natural cluster from the data set. It is unsupervised arrangement that means, it has no prearranged classes. This paper introduces analysis of many partitioning methodology of clustering algorithms and their relative research by reflecting their profits freely. Implementation of cluster analysis are Economic Science, Document categorization, Pattern Recognition, Image Processing, text mining. No particular algorithm is effective enough to solve problems from different fields. Hence, in this study some algorithms are presented which can be used according to one’s necessity. In this paper, different familiar partitioning- based methods – k-means, k-medoids and Clarans – are studied and compared. The study given here survey the behavior, nature and efficiency of these three methods. Data Mining is a method of classifying valid, suitable, new, logical pattern in the data.
Clustering, K-Means Algorithm , K-Medoids Algorithm , Clarans Algorithm