I have a dataset that contains name and organization pairs, and I'm trying to find a method that will either fill in edges and/or else cluster people based on common connections.
The edges sometimes connect people and other times connect people with an organization.
So, for example, maybe I have 'edges' that look like this:
Joe, OrgA Bill, OrgA Scott, Bill Scott, OrgB ... where usually the first column is a person, but the second column can be either a person or an organization.
My goal is to find how Scott is connected to OrgA (via Bill) and how Scott is connected to Joe (via Bill and OrgA), and how OrgA and OrgB are connected (via Scott and Bill).
I realize that maybe some data munging is needed to perhaps sort out the people and organizations, etc., but I was hoping I could find network analysis package that would cut down on some of the work for me.
What would you suggest?
I'm currently using igraph in R, but I've only just started and I'm not wedded to it. If anyone can recommend a types of approaches (as opposed to software packages), that would be fine too.
[link][comment]