Discussions of the Social Network Importer for NodeXL can also be found on the Codeplex site for that project:
The Facebook Fan page importer creates several types of networks from connections that form when two users like or comment on the same post.
From these edges several networks can be extracted:
User to post
User to user
Post to post
User to post is a bi-modal network in which multiple messages are linked to by different people who like or comment on that item. People and posts are two classes of object in that network. People can link to posts, but posts never link to people.
When a user likes or comments on more than one message they link them, indirectly, together. Visualizing these networks can offer a useful overview of the most recent 100 posts and the ways they group together and attract more or less attention
(likes) and commentary.
The indirect network created by people linking two messages can be made explicit. We can transform a user to post network into a post to post network. The Social Network Importer for NodeXL offers to deliver this network. A post-to-post network
reveals the clusters of messages that share a common appeal or interest to a group of users. Content categories can often be derived from these clusters. This network is reduced from a bi-modal network of people and messages to a unimodal network
that contains just messages (the people are present only in the form of edges).
The inverse of the post-to-post network is the user to user network. This is a network that is also derived by transforming the user to post network. Now users are linked to other users if they both liked or commented on the same messages. This
is also a unimodal network, there are only people (user accounts) left in the network. Messages are present only in the form of edges which link people. The user to user network is a useful way to discover groupings or clusters of users. Some fan
pages attract distinct social groups (ex: employees and customers), which becomes very visible in a user to user network.
I am not aware of any fan page network related publications (yet!). The data provider is still relatively new (report bugs as you find them, please!) and research with it is just starting. If anyone knows of Facebook Fan page research results
I would be happy to hear about them!