|
|
Visualizing with Paper Machines
Paper Machines, an add-on for the bibliographic management software Zotero, incorporates a range of text analysis tools into your web browser. There are several advantages to working with Paper Machines for text analysis, so we experimented to see if its various tools could help us hone our research questions about the Lincoln sermons in any meaningful way.
To create a body of texts for Paper Machines to examine, we first created a Zotero library. (If you’re unfamiliar with Zotero, check out this quick start guide.) Populating the Zotero library with the Lincoln sermons was a simple as pulling up each sermon from the Beck Center’s digital repository and clicking the “capture” icon in my navigation bar. Paper Machines stores the metadata for each sermon and will include the body of the sermons in the analysis.
If you’d like to try it for yourself, we’ve made the Zotero library publicly available at http://www.zotero.org/groups/lincolnsermons.
Paper Machines is easy to install and run, but if you’ve never done it before, this blog post will walk you through the process. First, let’s take a look at one of the word cloud:
The cloud raises a couple of potential points of inquiry. First, why aren’t the words “slave” or “slavery” prominent? Apparently, even though the vast majority of sermons in the collection are from the Northeast, slavery and emancipation aren’t necessarily among the more popular subjects. Secondly, the overall prevalence of the word “government” appears deserving of further investigation.
Let’s take a look at a couple of other modes of analysis integrated into Paper Machines. DBpedia annotation produces a web of related terms with clickable links:
DBpedia annotation. It’s aesthetically interesting, but it doesn’t necessarily contain anything surprising, nor does it provide more help than the word cloud.
Paper Machines also generates a host of phrase nets, which show the words on either side of common conjunctions. Some of these—”death the assassin;” “suffer the penalty”—seem like interesting places to look. Zeroing in on the sermons containing the terms is easy using the search function.
Phrase net showing words connected with “and.”
Other tools available using Paper Machines are MALLET topic modeling charts, a heatmap, and a flight path map. For this project, none of these proved particularly useful. Ultimately, 57 sermons on the same subject, all of which were delivered within a short number of months, did not provide enough variation tools to make many of the tools useful.