Founded in England in the mid-seventeenth century, the Quakers were Protestant Christians who dissented from the official Church of England and promoted broad religious toleration, preferring Christians’ supposed “inner light” and consciences to state-enforced orthodoxy. What are the subgroups and communities in the network?īefore there were Facebook friends, there was the Society of Friends, known as the Quakers.Who are the important people, or hubs, in the network?.What is the overall structure of the network?. This tutorial will help you answer questions such as: Insofar as even the most perceptive of scholars has difficulty perceiving, say, the overall shape of a network (its network “topology”) and identifying the nodes most significant for connecting groups, quantitative network analysis offers scholars a way to move relatively fluidly between the large scale social object (the “graph”) and the minute particularities of people and social ties. Factors such as their structural relation to further people and whether those additional people were themselves connected to one another have decisive influence on events. As sociologist Mark Granovetter pointed out in his important 1973 article “ The Strength of Weak Ties,” it’s rarely enough to notice that two people were connected with one another. Networks have long interested researchers in the humanities, but many recent scholars have progressed from a largely qualitative and metaphoric interest in links and connections to a more formal suite of quantitative tools for studying mediators, hubs (important nodes), and inter-connected structures. Check out the Programming Historian tutorials on installing Python and working with pip for more information. For this reason, when accessing Python 3 you will often have to explicitly declare it by typing python3 and pip3 instead of simply python and pip. It’s possible to have two versions of Python (2 and 3) installed on your computer at one time. Installed Python 3, not the Python 2 that is installed natively in Unix-based operating systems such as Macs (If you need assistance installing Python 3, check out the Hitchhiker’s Guide to Python) and.a basic familiarity with networks and/or have read “From Hermeneutics to Data to Networks: Data Extraction and Network Visualization of Historical Sources” by Martin Düring here on Programming Historian.You’ll likely want a combination of visualization and network metrics in your own project, and so we recommend this article as a companion to this earlier Programming Historian tutorial. We will therefore focus on ways to analyze, and draw conclusions from, networks without visualizing them. N.b.: This is a tutorial for exploring network statistics and metrics. To analyze humanities network data to find:.To use the NetworkX package for working with network data in Python and.Advanced NetworkX: Community detection with modularity.What might you learn from network data?.This, similar to the degree centrality method, returns a dictionary of node: betweenness centrality value pairs. To get the betweenness centrality of a network-graph, we simply use 'nx.betweenness_centrality(network-graph)'. Concerning a bridge between conservative and liberal clusters of YouTube accounts, I would be willing to bet that YouTuber 'Philip DeFranco' has a high betweenness centrality number for political YouTubers, because both parties of the political spectrum watch his show. Whereas degree centrality shows how connected a node is, betweenness centrality reveals 'bottleneck' or 'bridge'-like nodes, as would say, that reveal links between different clusters within a network. If a node has a betweenness centrality number of '0', that would imply it has no shortest paths, meaning it's a very inefficient way to disseminate information from that node to other parts of the network. This is the number of shortest paths a node actually has divided by all the potential shortest paths in a network. Twitter node '0' had no edges, thus explaining the zero value for its degree centrality.Īnother way to describe a node's importance, or usefulness, rather, is through a measure called 'betweenness centrality'. (we're ignoring directionality with edges in this example). Hypothetically, if these nodes represented users on Twitter, and node 11 was John Mayer and node 8 was Charlie Carrera, me and John Mayer would have the same degree centrality values the fraction of our followers divided over all potential followers would be the same. This graph method returns a dictionary where the keys are the nodes and the values are the degree centrality estimates.
0 Comments
Leave a Reply. |