Respond to JDH 2.1

The Journal of Digital Humanities “catches the good” by surfacing and highlighting valuable material published on the open web. For this issue we will go one step further, and solicit responses to this edition’s take on topic modeling in the digital humanities.

We offer two ways to provide your response. On this page you can post an open comment or a link to longer piece posted elsewhere. You also can tweet a response using #JDHTopics, which we will aggregate on another page.

We will then provide a summary and analysis of the comments, using both topic modeling and close reading.* You can point to work made before the publication of this issue, but our own topic modeling of the comments will be limited to pieces created between April 11 and May 11.

We have two goals with this open call for comments. First, we want to give the community a space to really discuss and engage with the content of the issue. Second, we want to provide readers the chance to see the topic modeling of a corpus they are helping to create. Ultimately, we hope to help readers gain a better understanding of this somewhat opaque method.

*The responses did not provide a large enough corpus for topic modeling, however they remain here for you to read in full.

 

Leave a comment

Get a Trackback link

2 Trackbacks/Pingbacks

  1. Pingback: Topic Modeling in the Humanities ← the scottbot irregular on April 11, 2013
  2. Pingback: Journal of Digital Humanities on April 15, 2013

2 Comments

  1. This issue of JDH is great, and it got me thinking about some of the lessons I have learned over seven or eight years of topic modeling literary corpora. I thought it might be useful to readers of this issue of JDH if I posted a few of those lessons online. My "secret" topic modeling recipe is now available at http://www.matthewjockers.net/2013/04/12/secret-r

  2. In addition to the already-excessive two appendices to my article, I put up a blog post to include as more grist for the mill.

    I also have this temptation to run Dunning log-likelihood or some basic comparison algorithms over the pre- and post-revisions texts.