I have N graphs, they all have same topology,but they differ in vertices weight. I would like to cluster these graphs such that graphs that their vertices has similar weights in similar positions are in same cluster. Initially I was thinking to vectorize graphs(every element of vector is one vertice) and compute euclidian distances of them. But this does not take the topology of graph into account Would someone can help me with this?
Asked
Active
Viewed 90 times
1 Answers
0
You could vectorize the graph and do co-ordinate wise clustering i.e., apply Euclidian distance clustering for each coordinate
Sid
- 2,617
-
1would you please explain more what do you mean by that. – user2806363 Oct 15 '17 at 07:42