The Uncertainty of Influence: Do Books Have DNA?

Today, in the Stanford Humanities Center Board Room, as part of the Visualizing Uncertainty workshop that Nicole Coleman has organized, I’ll be presenting on using network analysis and visualization in the analysis of literature.

Visualizing Uncertainty in Literary Influence Networks

Matthew Jockers and Elijah Meeks
Monday, December 5, 1:30pm
Stanford Humanities Center, Board Room

About this Session:
Using tools and techniques from natural language processing, text-mining, and social network analysis, we analyze a corpus of 4,300 19th-century novels in order to detect and track stylistic and thematic change over time. Our talk will emphasize the confounding factors that affect our examination: the creation of our corpus, our metric for measuring similarity, and the visual representation of the data as a network.

This entry was posted in Text Analysis, Visualization. Bookmark the permalink.

2 Responses to The Uncertainty of Influence: Do Books Have DNA?

  1. This looks like a fabulous talk. I’ve been watching the stunning video that was uploaded to YouTube of the “Nineteenth-Century Literary Genome.” I’d like to share the video with some of my colleagues here at Penn, but I have a couple of questions. First, was the video produced using Gephi? And second, what works/genres are in that strange polyp that explodes from the side of the cluster? It’s a truly astounding visualization.

    • Elijah Meeks says:

      Thanks, Devin. To answer your questions:

      The animation was built with code I wrote for Gephi Toolkit. I claimed I was releasing this code a couple weeks back, and I still haven’t, so I really need to spend some time cleaning it up and then I’ll post it on Github.

      As far as any questions of genre, authorship, similarity, et cetera, you should direct those to Matt Jockers. I only deal with the interesting, attractive and fun parts of these problems, and none of the tedious bits.