zme1 / toscana

A repository to house research and web development for the Lega Toscana project, led by professor Lina Insana (Spring 2018) and professor Lorraine Denman (Fall 2018), and with consultation from members of the DH Advanced Praxis group at the University of Pittsburgh at Greensburg.
http://toscana.newtfire.org
3 stars 1 forks source link

Cytoscape data preparation #12

Closed zme1 closed 6 years ago

zme1 commented 6 years ago

@ebeshero I'm done (for now) with all of my document analyses, and today I'll be starting my work on social network mapping. I wouldn't say I'm confused by the introduction material as much as I simply want to verify that my conceptualization of it is correctly adapted for my work...

I think the most prominent and interesting social network will be that which visualizes proposals (<seg> of @type="proposal" in my corpus), and I want to be sure that I am processing all my information in the correct places...

My nodes will obviously be the two (or more) participants in the interaction -- since the interaction is more directional than non-directional (although I still think it does not cleanly ascribe to either), I think I'll treat it as a directional one to see who proposes more versus who supports proposals more.

The edge will actually be the attribute value of <seg>, which is, in this context, "proposal". I am planning to revisit all my proposals in the corpus and specifying whether the proposal is fiscal, social, or regulatory once I get a head start on my site and analyses, but once those @subtype attributes are attached, they will be considered edge attributes for the map.

Since I'll be treating the proposals as directional interactions, that would mean that the node attributes would distinguish proposers from supporters in my data set.

If I'm understanding this correctly, my first import will only be the participant names, and the edge that connects them, without any attribute data. So it will simply consist of raw connections between each of the members of the Legal, without taking into consideration how they're interacting just yet...