Credential Clustering in Parallel Comparability Frequency Amplitude

Cluster analysisorclusteringis the process ofassigning a set of objects into groups such that theobjects in the same cluster are more similarto eachother and objects in other clusters are dissimilar. Inthe existing system, Locality preserving indexingmethod is used. Euclidean distance is used as themethod to find the distance between the documents.This method also takes use of weightedfunction tofind the distance between documents. Majordifficulty is found in analyzing the weightedfunction. Since the documents can take manyforms, the structure of the document space dependson the similarities between the documents. A newapproach based on Correlation as a similaritymeasure is best proposed for analyzing the structureof documents arranged in the high-dimensionaldocument space