Get the distance to a node or set of nodes

Post date: May 28, 2014 9:57:25 PM

Suppose you want to measure, for each node in the network, how many links away they are from the three most important nodes. How do you do it? First, you need to make a list of the important nodes. Suppose the dataset is campnet and the nodes are Pat, Steve and Russ. Now, you need to compute geodesic distances. In the ucinet command line (matrix algebra), type:

->geo = geodesic(campnet)

This tells you how many links each person is away from everyone else. Now we need to narrow it down to Pat, Steve and Russ. We do this:

->y = col(geo pat steve russ)

So Y has the distances from each person to each of the key nodes. Next, we have to decide what exactly we want -- the average of a person's distance to pat, steve, and russ? or perhaps the minimum? Let's assume minimum. Type:

->z = wmin(y row)

->dsp z

The vector Z gives the distance from each person to the closest of the three key players. We could have chosen the average instead:

->z = wavg(y row)

Note that these are distances from a node to the key nodes. In many cases, this might not be as interesting as distances to a node from the key nodes. To do that, we would simply start the process by calculating the transpose of the distance matrix:

-->geo = transpose(geodesic(campnet))