eng Ansari Education and Research Society Journal of Ultra Scientist of Physical Sciences August 2010 22 2 629 641 jusps-A 1014 article S. Gountia (sujata_attabira@rediffmail.com) 1 S. K. Sahoo (sahoosk1@rediffmail.com) 2 Teacher Fellow, G.M. College, Sambalpur (INDIA) Orissa Institute of Mathematics & Applications, Bhubaneswar (INDIA) ORISSA <p style="text-align: justify;">M. R. Anderberg (1973) and B. S. Everitt (1993) have developed the cluster Analysis which is a process to decompose the set of objects from a finite set <img src="file:///D:/atif/Drive%20E%20data/desktop%20data/Business/journal/JUSPS%20A&amp;B/JUSPS/22(2m)/AbsWMF15.JPG" style="height:11px; width:13px" />into subgroups or clusters based on similarity. Crisp clustering techniques are classified into 3 groups according to the algorithmic approach as (i) Hierarchical clustering method (ii) Graph-Theoretic clustering method (iii) Objective function based clustering method. Graph-theoretic clustering methods are normally based on some kind of connectivity of the nodes of a graph representing the data set. The fuzzy graph approach is more powerful in cluster analysis than the usual graph-theoretic approach. The concept of fuzzy graph with edges associated with two types of weights were studied before⁹. In this study a min-max weight of the cut set and min-max edge connectivity are introduced. The clustering technique of narrow slicing procedure² based on edge connectivity has been adopted for determining -edge components of a fuzzy graph.</p> <p>&nbsp;</p> <p>&nbsp;</p> https://www.ultrascientist.org/paper/1014/ Fuzzy graph cluster connectivity edge connectivity vertex connectivity