Performance Evaluation with K-Mean and K-Mediod in Data Mining

Data mining is the process of extraction of various types of information from different types of dataset that contains various types of attributes. Clustering is an approach that divides the whole information into different clusters. After processing of division of data values into different clusters centeroid have been computed. Cluster centeroid has been done on the basis of distance from other cluster members available in the particular clusters. The main problem in the clustering for data mining process is that text mining contains different problem for division of the text dataset into different cluster. Sometimes in the process of clustering by default empty cluster has been developed. We removed this problem by using K-mean clustering with hybridization of K-mediod algorithm.