Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Learning Bayesian Networks
Hi-index | 0.00 |
This paper presents new idea for Markov blanket approximation. It uses well known heuristic ordering of variables based on mutual information, but in another way then it was considered in previous works. Instead of using it as a simple help tool in a more complicated method most often based on statistical tests - presented here idea tries to rely without any further statistical tests only on the heuristic and its previously not considered interesting properties.