Understanding the content of a large text corpus can be assisted by topic modeling methods, but the discovered topics often do not make clear sense to human analysts. Interactive topic modeling addresses such problems by allowing a human to steer the topic model curation process (generate, interpret, diagnose, and refine). However, human have limited ability to work with the artifacts of computational topic models since they are difficult to interpret and harvest. This paper explores the nature of such challenges and provides a visual analytic solution in the context of supporting political scientists to understand the thematic content of online petition data. We use interactive topic modeling of the White House online petition data as a lens to bring up key points of discussions and to highlight the unsolved problems as well as potentials utilities of visual analytics methods.
All Science Journal Classification (ASJC) codes
- Computer Science(all)