
If I have a matrix with 'closeness' data, i.e. each member of a group has specified on a scale of 010 how much they interact with someone, how might I make use of that? For example, I might have a matrix that resembles:
A B
C D
A 10
5 1
B 8
6 6
C 4
7 2
D 1
5 5
I would like to use these data to better inform my graph, rather than simply resorting to defining vertices by an arbitrary cutoff of their closeness scores.
Thanksthis seems like a great tool and I'd love to be able to use it for this case!



Well, one thing you could do is set the width of the connecting edges to be proportional to the closeness scores, which will make the close connections stand out in the graph. Assuming that you have your matrix stored in a standard Excel workbook,
you would do the following:
1. Use NodeXL, Data, Import, From Open Matrix Workbook to import the matrix into NodeXL's Edges worksheet. This will create an Edge Weight column that contains the closeness scores from the matrix.
2. Use NodeXL, Visual Properties, Schemes to set the edge widths. Select the Weighted Graph Scheme, and set Weight Column to Edge Weight.
If, instead, you wanted to base edge color or edge opacity on the closeness scores, you could use NodeXL, Visual Properties, Autofill Columns instead of Schemes.
Let me know if you need more details.
 Tony

