A cluster is grouping of similar objects. In
clustering we find object that must be sufficiently close (or
similar) to one another. In this paper we deal with a new graph
structure called metagraph, which show meta-node to meta-node
mapping. This paper explains the clustering methods, based on
metagraph clustering. The metagraph clustering method based
on the concept that intra-cluster and inter-cluster similarity.
We use concept of clustering to construct a metagraph.
Clustering metagraph is a natural way for handling
Published In : IJCSN Journal Volume 3, Issue 5
Date of Publication : October 2014
Pages : 402 - 404
Figures : 01
Tables : --
Publication Link : Formation of Metagraph Using Clustering
Metagraph allows different
components of the process to be represented both
graphically and analytically. Considering an example
which show a metagraph which is constructed in matlab
using some function of matlab . Using the same we also
construct a metagraph for social data clustering and make
a social metagraph.
 Alsaleh Slah, Nayak Richi and Xu Yue, Grouping
People in Social Networks Using a Weighted Multi-
Constraints Clustering Method, WCCI 2012 IEEE
World Congress on Computational Intelligence . 2012.
 Basu A. and Blanning R. W. ,Metagraph , Omega
International Journal of Management Sciences, Vol.23,
1995, pp. 13-25.
 C.T Zahn(1971) : Graph-theoretical methods for
detecting and describing clusters , IEEE transaction on
 Gaur D., Shastri A. and Biswas R. , Metagraph: A new
model of data structure ICCSIT IEEE Computer Society
Singapore, 2008, pp.729-733.
 Lahiri and T. Berger-Wolf, Mining Periodic Behavior
in Dynamic Social Networks,Proceedings of the 8th
IEEE Inter national Conference on Data Mining, 2008,
 M.E.J. Newman and M. Girvan, Finding and
Evaluating Community Structure in Networks,Physical
Review E, vol. 69, no. 2, 2004, pp. 026113-15.
. M.E.J. Newman, Fast Algorithm for Detecting
Community Structure in Networks,Physical Review E,
vol. 69, 2004, pp. 066133-1.
 Thair Nu Phyu, Survey of Classification Techniques in
DataMining IMECS 2009, Hong Kong.