Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Learning belief networks from data: an information theory based approach
CIKM '97 Proceedings of the sixth international conference on Information and knowledge management
Machine Learning - Special issue on learning with probabilistic representations
Applying general Bayesian techniques to improve TAN induction
KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Bayesian Network Mining System
Proceedings of the International Symposium on "Intelligent Information Systems X"
Learning with mixtures of trees
The Journal of Machine Learning Research
Feature subset selection by genetic algorithms and estimation of distribution algorithms
Artificial Intelligence in Medicine
Approximating discrete probability distributions with dependence trees
IEEE Transactions on Information Theory
Consistency of an estimate of tree-dependent probability distributions (Corresp.)
IEEE Transactions on Information Theory
Hi-index | 0.00 |
Bayesian networks have many practical applications due to their capability to represent joint probability distribution in many variables in a compact way. There exist efficient reasoning methods for Bayesian networks. Many algorithms for learning Bayesian networks from empirical data have been developed. A well-known problem with Bayesian networks is the practical limitation for the number of variables for which a Bayesian network can be learned in reasonable time. A remarkable exception here is the Chow/Liu algorithm for learning tree-like Bayesian networks. However, its quadratic time and space complexity in the number of variables may prove also prohibitive for high dimensional data. The paper presents a novel algorithm overcoming this limitation for the tree-like class of Bayesian networks. The new algorithm space consumption grows linearly with the number of variables n while the execution time is proportional to n·ln(n), hence both are better than those of Chow/Liu algorithm. This opens new perspectives in construction of Bayesian networks from data containing tens of thousands and more variables, e.g. in automatic text categorization.