In 2003 David M. Blei, Andrew Y. Ng, and Michael I. Jordan presented Latent Dirichlet Allocation (LDA) as a graphical model for topic modeling in natural language processing.
Blei, Ng, & Jordan. "Latent Dirichlet Allocation," The Journal of Machine Learning Research, Volume 3 (2003) 993-1022.
In 2013 Blei published an informative illustrated survey of algorithms for managing large document archives: "Probablilistic Topic Models," Communications of the ACM Vol. 55, No. 4 (2012) 77-84, accessed 10-20-2013.