Recently I’ve been looking at a couple of projects in computational linguistics, which harks back to my days at SpeechBot and AgileTV. There has been a lot of interesting research in the field since then, particularly in the area of nonparametric Bayesian inference. The homepage of David Blei is an excellent resource, including an annotated bibliography and links to a variety of software implementations.
The following videos also provide a nice overview:
Modeling Science: Dynamic Topic Models of Scholarly Research
A/Prof David M. Blei, Princeton
Transparency and Topic Models
Aspro Hanna M. Wallach, UMass Amherst