"Networks are everywhere" is THE catch phrase of social network analysis. As I said in a previous post, if you are a network scientist everything looks like a network to you. That's why I decided to start a series about "everywhere" networks. But not just any. No! In particular those that are just not interesting at all. Although networks might be ubiquitous, that does not always make them worthwhile studying.
With this special kind of networks, I will just do some fancy visualizations and maybe some boring statistical analysis. To emphasize the non-scientific relevance, the visualizations and statistics will be presented with the dearly beloved comic sans.
Body Part Relationships Network
I got the data from here
. The description of the data is as follows:
"[...] contributors classified if certain body parts were part of other parts. Questions were phrased like so: "[Part 1] is a part of [part 2]," or, by way of example, "Nose is a part of spine" or "Ear is a part of head."
Well if that's not exciting...
Here is a visualization of the network
A cluster analysis revealed, to which class of body parts (upper in light blue, lower in light red) certain body parts belong. The algorithm was not so sure about the belly though.
The following table shows the most important body parts according to some centrality measures and what these measures could stand for.
So the head is the most important body part and the face is full of other things. Good to know.
Labels: Network Analysis, Network Visualization, NotInteresting