Editor/Reviewer loginFrequency : 12 issues per year
Subject : Computer Applications and Technology
ISSN : 2319–8656 (Online)
IJCATR Volume 5 Issue 4
Usage of Self-Organizing Map for Clustering Vertices
Rashmi Lad P S Metkewar R.S. Walse
10.7753/IJCATR0504.1005
keywords : Test automation, self-organizing map, topology, visualization, clustering, Euclidean distance.
Usage of Self-Organizing Map (SOM) for clustering vertices of any given graph. Simultaneously its input is observed and worked in terms of weight matrix, learning rate and final resultant matrix, which helps to form a cluster. The purpose of this paper is to introduce a procedure of SOM for clustering and observed impact corresponding to varied weight class for simple graph or vector matrix using Euclidean distance. A simple vector matrix problem is solved by using 2, 3 & 4 weight class matrix. By adopting a different weight matrix class with same vector matrix has presented a clustering and visualization.
@artical{r542016ijcatr05041005,
Title = "Usage of Self-Organizing Map for Clustering Vertices",
Journal ="International Journal of Computer Applications Technology and Research(IJCATR)",
Volume = "5",
Issue ="4",
Pages ="202 - 209",
Year = "2016",
Authors ="Rashmi Lad
P S Metkewar
R.S. Walse"}
The self-organizing Map is a special type of neural network that accepts N-dimensional input vector and maps them.
To introduce a procedure of SOM for clustering and observed impact with different weight matrix.
The topological neighboring decline monotonically, from a value less than half the largest diagonal of the map.
To explain Self-Organizing Map (SOM) for clustering vertices and their usage.