Academic Journals Database
Disseminating quality controlled scientific knowledge

ANALYSIS OF CLIQUE BY MATRIX FACTORIZATION AND PARTITION METHODS

ADD TO MY LIST
 
Author(s): Raghunath Kar | Dr. Susant Kumar Das

Journal: International Journal of Computer Science and Management Studies
ISSN 2231-5268

Volume: 11;
Issue: 03;
Start page: 09;
Date: 2011;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: CLIQUE | APRIORI | dense unit

ABSTRACT
In real life clustering of high dimensional data is a big problem. Tofind out the dense regions from increasing dimensions is one of them.We have already studied the clustering techniques of low dimensionaldata sets like k-means, k-mediod, BIRCH, CLARANS, CURE, DBScan, PAM etc. If a region is dense then it consists with number of data points with a minimum support of input parameter ΓΈ other wise itcannot take into clustering. So in this approach we have implementedCLIQUE to find out the clusters from multidimensional data sets. Indimension growth subspace clustering the clustering process start atsingle dimensional subspaces and grows upward to higher dimensionalones. It is a partition method where each dimension divided like a grid structure. In this paper the elimination of redundant objects from the regions by matrix factorization and partition method are implemented. The comparisons between CLIQUES with these two methods are studied. The redundant data point belongs to which region to form a cluster is also studied.
Save time & money - Smart Internet Solutions     

Tango Rapperswil
Tango Rapperswil