Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Rough Sets: Theoretical Aspects of Reasoning about Data
Rough Sets: Theoretical Aspects of Reasoning about Data
A Guide to the Literature on Learning Probabilistic Networks from Data
IEEE Transactions on Knowledge and Data Engineering
Technologies for constructing intelligent systems
| Hi-index | 0.00 |
Bayesian network (BN) is a directed acyclic graph encoding probabilistic independence statements between variables. BN with decision attribute as a root can be applied to classification of new cases, by synthesis of conditional probabilities propagated along the edges. We consider approximate BNs, which almost keep entropy of a decision table. They have usually less edges than classical BNs. They enable to model and extend the well-known Naive Bayes approach. Experiments show that classifiers based on approximate BNs can be very efficient.