Richard E. Neapolitan

Just starting to work my way through this. It looks to be a good textbook on Bayesian networks, accessible to anyone with a basic foundation in probability and graph theory.

A similar book is Judea Pearl's Probabilistic Reasoning in Intelligent Systems.

Neapolitan's book focuses exclusively on directed graphical models (aka Bayes nets). See rel="nofollow" target="_top">Pearl's book if you want a discussion of undirected graphical models (aka Markov nets) as well.

There is a discussion of Markov Chain Monte Carlo methods, but it assumes some familiarity with the topic. I found a better introduction to this subject to be the first article in Markov Chain Monte Carlo in Practice.

Errata:

• Definition 2.8 in Section 2.2 on page 94: "Let gp be a dag pattern whose nodes are the elements of V, and A, B, and let C be mutually disjoint subsets of V" should read "...whose nodes are the elements of V and let A, B, and C be mutually disjoint subsets..."


• Section 2.3.1, page 97: "For example, in Figure 2.18 (a), because there is no edge between V and C..." should read "...between V and S..."

… (más)