Books about US Politics: Network Visualization

Network of books about US politics with its frequent copurchasing of books by the same buyer.

Now, I am working on the project at the Knowledge Lab about books and its political spectrum from conservative to liberal. And I found this publicly available dataset online, so I think it would be good for me to try to visualize this network and see the relationship between books and its copurchase.

I got this dataset from Prof. Mark Newman's personal website page of Network data (If you are interested, you can check his website as well.) He is a physics professor at University of Michigan, conducting research on the structure and function of social and information networks. According to Prof. Newmann's website, this dataset contains

A network of books about US politics published around the time of the 2004 presidential election and sold by the online bookseller Amazon.com. Edges between books represent frequent copurchasing of books by the same buyers. The network was compiled by V. Krebs and is unpublished. 

Basically, I used Gephi to visualize this dataset. The nodes are books about US politics published around 2004 and the edge is the degree of co-purchasing of books by the same customer. Kerbs also manually labeled each book for their political spectrum whether it is conservative (here it's represented by red), neural (green) or liberal (blue) based on synopsis and reviews about the books. The size of the node (and also the size of the label) corresponds to the number of books that are copurchased with this particular books, and the degree of the edge (which is not represented in this visualization) is how many times two books are copurchased together. 

Cluster on Conservative side

Cluster on Liberal side

From this network, we see that books from the same political spectrum are more likely to be copurchased together, implying that people with a certain ideology trend to emphasize their own believes by consuming similar media (in this case books) with the same ideology of their own.

The same phenomenon are happened everywhere. For example, liberal people tends to follow liberal thinkers and writers on Twitter while conservative people tend to follow conservative ones. By having basically unlimited amount of information, the internet allows people to reemphasize their own belief by choosing to consume media with a certain ideology. Whereas before the internet, people will have less opportunity to pick media they consume because there is less information a person can obtain. We can say that this made people less open about other ideologies and overconfident about their own believes. Therefore, we should always consciously pick media we consumed. By diversifying the source of information, we will get a better picture of each topic and what other people think about the issue and why they think it that way.