Comments on: Demystifying Networks

By: Developing a Workshop in Network Analysis for Humanities – DH @ The Queen's University Library

Sun, 13 Mar 2016 17:49:37 +0000

[…] Demystifying Networks – Scott Weingart […]

By: Explore The Historian’s Digital Toolbox | exodus777blog

Explore The Historian’s Digital Toolbox | exodus777blog — Thu, 10 Mar 2016 01:43:08 +0000

[…] Weingart, Scott, “Demystifying Networks,” scottbot irregular, December 14, 2011, http://www.scottbot.net/HIAL/?p=6279 […]

By: Palladio and Google Fusion Tables Mapping | Introduction to Digital Humanities

Tue, 10 Nov 2015 20:41:43 +0000

[…] more things, represented by nodes and edges, as explained by Scott Weingart in his article, “Demystifying Networks.” In the example I used, the data set was of primarily the relationships of Ralph Neumann to […]

By: Citation Networks and Disciplinary Histories | FSU Digital Scholars

Citation Networks and Disciplinary Histories | FSU Digital Scholars — Sun, 04 Oct 2015 08:06:58 +0000

[…] Weingart, “Demystifying Networks” (Dec. 14, 2011), “From Trees to Webs: Uprooting Knowledge through Visualization” […]

By: Learning about Networks | Caitlin Chaves Yates

Learning about Networks | Caitlin Chaves Yates — Wed, 13 May 2015 08:46:53 +0000

[…] in an attempt to learn more about how networks work. So far I’ve found Scott Weingart’s Networks Demystified series to be super helpful for an introduction. I’ve also made good use of Shawn […]

By: Data Visualization and Network Analysis | stokesj8

Data Visualization and Network Analysis | stokesj8 — Mon, 20 Oct 2014 14:48:10 +0000

[…] tools in the DH field when used appropriately and correctly. Like Scott Weingart said in his Demystifying Networks blog, “networks can be used on any project. Networks should be used on far […]

By: Christof

Christof — Fri, 05 Sep 2014 15:33:04 +0000

Hi Scott, thanks for this very clear and to-the-point post. When I recently experimented with making a network of crime fiction novels and their topics, I had to deal precisely with some of the issues you discuss: how to deal with a bimodal network and how to deal with high density.
As for the bimodality, for example, depending on the relative number of topics and novels, topics may just on account of their smaller number have higher centrality, which I guess would then be a misleading measure. In my case, even with a ratio of 200 novels to 40 topics, this is certainly noticeable.
And as for the sparsity issue, cutting out a relatively large part of the low-scoring novel-topic associations certainly helped with visualization. But it seems choosing the cut-off point is a matter of trial-and-error and experience. In my case, I took out any novel-topic associations with a score lower than 0.05, of which there were many, reducing the number of edges by about two thirds. But this will vary with the number of documents and number of topics in the dataset.
Anyway, this post is still a good read!

By: Scott Weingart

Scott Weingart — Mon, 23 Sep 2013 19:17:54 +0000

Dan, thanks for the comment, that’s a great point, I’ll update accordingly in my next write-through.

By: Daniel Shore (@DanAShore)

Daniel Shore (@DanAShore) — Mon, 23 Sep 2013 18:29:08 +0000

Hey Scott,
This is a really helpful, clear, post on what networks do and don’t do well. One thing that you might consider clarifying is what you mean by “useful,” as when you say that “dense networks networks are rarely useful.” I gather that by “useful” you mean “susceptible to algorithmic analysis.” But of course there are plenty of other things for which networks are useful.
d

By: Networks Demystified 5: Communities, PageRank, and Sampling Caveats | the scottbot irregular

Wed, 18 Sep 2013 20:59:32 +0000

[…] installment of Networks Demystified will be a bit more applied than the previous bunch (1 network basics, 2 degree, 3 power laws, 4 co-citation analysis). Like many of my recent posts, […]