Not Logged In

Learning Bayesian Networks from Data: An Information-Theory Based Approach

This paper provides algorithms that use an information-theoretic analysis to learn Bayesian network structures from data. Based on our three-phase learning framework, we develop efficient algorithms that can effectively learn belief networks, requiring only polynomial numbers of conditional independence (CI) tests in typical cases. We provide precise conditions that specify when these algorithms are guaranteed to be correct as well as empirical evidence (from real world applications and simulation tests) that demonstrates that these systems work efficiently and reliably in practice.

Citation

J. Cheng, R. Greiner, J. Kelly, D. Bell, W. Liu. "Learning Bayesian Networks from Data: An Information-Theory Based Approach". Artificial Intelligence (AIJ), 137(1-2), pp 43--90, January 2002.

Keywords: belief nets, machine learning, information theory
Category: In Journal

BibTeX

@article{Cheng+al:AIJ02,
  author = {Jie Cheng and Russ Greiner and Jonathan Kelly and David Bell and
    Weiru Liu},
  title = {Learning Bayesian Networks from Data: An Information-Theory Based
    Approach},
  Volume = "137",
  Number = "1-2",
  Pages = {43--90},
  journal = {Artificial Intelligence (AIJ)},
  year = 2002,
}

Last Updated: April 24, 2007
Submitted by Russ Greiner

University of Alberta Logo AICML Logo