04 January 2013

Unasked Questions About Network Analysis

As a retrospective, I thought I'd put all the previous networks up in one post to show how each of the Companions [ or Companion pairs ] compared. You'll notice that in each Companion's network, they are the closest character to the Doctor. After looking at all these networks together, I thought it might be helpful to answer a few unasked questions. Because I know that when I first started, they were the type of questions I had to ask.

At the end of the Second Doctor's network analysis I'm sure there will be additional unasked questions that I will think of.


Susan Friends

Barbara & Ian

Barbara Ian Friends


Vicki Friends


Steven Friends


Dodo Friends

Ben & Polly

Ben Polly Friends

First Doctor

01Doctor Friends

Why do you call him the "First Doctor" in the text but I see "Doctor Who 01" in the network? Just like you say "Susan" and it's "Susan Foreman" in the network?

All the Doctors are listed as Doctor Who 01, Doctor Who 02,... for sorting purposes. That way I have all the Doctors listed in the same place in the dataset. And I generally drop the surnames when writing about the characters unless the surname is useful [ such as in the case of lesser known characters ].

And in the case of a dataset like this, I am fairly confident that everyone will know exactly who I am talking about.

Why do you use those colours? Is there any significance?

No, it's all arbitrary and the colours are selected to work the best on the black background. If need be, I can select special colours for individual characters.

Does the position of the character mean something special?

Not really. Each time I start up the visualizer, the characters will take up slightly different positions on the display. They will all be at the same distances from each other but that's all. And occasionally in those complicated sections, the characters may assume different positions within that cluster. Again, it does not change the final strength.

Why, in the Doctor's network, is the closest character The Daleks? They weren't a Companion?

First off, you'll notice that both the Doctor and the Daleks are both connected to many of the perimeter characters on an fairly equal basis. No other two characters have that many links to other people. So, they are both placed in the middle. But it is the numbers that really matter.

The link strength between the Doctor and the Daleks is only [ 4 ] whereas the Doctor link strength to Susan is [ 12 ] and to Barbara & Ian [ 17 ]. The position of characters can, at times, be deceiving. That is why you can not go by the network picture alone. There's not one other network where the Daleks are closer to the Doctor than any of the Companions.

How can each Companion be the most important person in one network and a complete nothing in another?

Each characters' network is designed around a function called Sphere of Influence. When the network is created I start with one characters. Then I tell the program to find every character and story that is connected to the character. This will, sort of, make that character the most important person in their network. But that's not the only thing I look at.

The charts showed whether that character ranked far above everyone else or whether many other characters scored nearly as well or equal. Having many character score similarly decreases a characters potential power within their network. Dodo is an excellent example. 

No comments:

Post a Comment