Abstract

Algorithms for learning Bayesian networks from data havetwo components: a scoring metric and a search procedure. The scoring metric computes a score reflecting the goodness-of-fit of the structure to the data. The search procedure tries to identify network structures with high scores. Heckerman et al. (1995) introduce a Bayesian metric, called the BDe metric, that computes the relative posterior probabilityofanetwork structure given data. In this paper, we show that the search problem of identifying a Bayesian network---among those where each node has at most K parents---that has a relative posterior probability greater than a given constant is NP-complete, when the BDe metric is used. 12.1

Citations

8881	D.S.: Computers and Intractability, A Guide to the Theory of NP-Completeness - Garey, Johnson - 1979
877	A Bayesian method for the induction of probabilistic networks from data - Cooper, Herskovits - 1992
28	Learning and robust learning of product distributions - Höffgen - 1993
5	A transformational characterization of Bayesian network structures - Chickering - 1995
5	Learning discrete Bayesian networks - Heckerman, Geiger, et al. - 1995
2	Some NP-complete problems on graphs - Garvil - 1977
2	Learning and robust learning of product distributions - gen, U - 1993