Count of indirect edges

Mar 5, 2013 at 1:45 PM
Hi,
In my project I want to measure the indirect linkages between the vertex. So what I need is if lets say Vertex 1 is directly connected to 3 other vertices and those 3 are to another 2 each, I would like to know the break up of what is the direct and indirect connection for Vertex 1.

Please help me in knowing how to accomplish this. Is this something readily available inside the interface by means of changing some settings or I need to write some code etc.

Thanks & Regards
Mar 6, 2013 at 6:37 AM
I'm not sure what you mean by "the break up of what is the direct and indirect connection." Are you interested in the shortest paths between each pair of vertices in your graph? If so, NodeXL won't do that for you, but you can do it yourself in C# or VB.NET by implementing what's called "Dijkstra's algorithm." That algorithm is documented, complete with pseudocode, at http://en.wikipedia.org/wiki/Dijkstra%27s_algorithm.

-- Tony
Mar 6, 2013 at 11:42 AM
Hi,
What exactly I mean is :-
Lets say I am connected to you and you are connected to two more then my direct connection is 1 (i.e. you) and my indirect connection is 2 (i.e. your two direct connection).

I hope I am clear now, since I am a beginner as of now so dont know the technical terms to explain it better.
Coordinator
Mar 6, 2013 at 8:29 PM
Thanks for the interest in NodeXL!
It sounds to me that you are interested in a "2-degree" network and would like to know the Vertex count of that network for each person in your graph.
A 1-degree network directly connects a specific entity to other entities. A 1.5 degree networks adds the links among the other people. A 2 degree network includes links from people you connect to to people they connect to (even if you do not connect to them).
You can then export an 2-degree network via NodeXL and then run analysis on that network to measure its vertex count.

Regards,

Marc
Developer
Mar 7, 2013 at 2:48 AM
Continuing Marc's comment, you could use the subgraph images feature to show the 2-degree network for each node. Then you can get an idea for each node how large that subgraph is and what kinds of behavior it represents.