A Characterization of the Dirichlet Distribution with Application to Learning Bayesian Networks
Dan Geiger, David Heckerman
We provide a new characterization of the Dirichlet distribution. This characterization implies that under assumptions made by several previous authors for learning belief networks, a Dirichlet prior on the parameters is inevitable.
PS Link: http://www.research.microsoft.com/research/dtg/heckerma/TR-94-16.htm
PDF Link: /papers/95/p196-geiger.pdf
AUTHOR = "Dan Geiger
and David Heckerman",
TITLE = "A Characterization of the Dirichlet Distribution with Application to Learning Bayesian Networks",
BOOKTITLE = "Proceedings of the Eleventh Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-95)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "1995",
PAGES = "196--207"