The topological fusion of Bayes nets
UAI '92 Proceedings of the eighth conference on Uncertainty in Artificial Intelligence
Modeling and Reasoning with Bayesian Networks
Modeling and Reasoning with Bayesian Networks
Hi-index | 0.00 |
This paper deals with the problem of merging multiple-source uncertain information in the framework of probability theory. Pieces of information are represented by probabilistic (or bayesian) networks, which are efficient tools for reasoning under uncertainty. We first show that the merging of probabilistic networks having the same graphical (DAG) structure can be easily achieved in polynomial time. We then propose solutions to merge probabilistic networks having different structures. Lastly, we show how to deal with the sub-normalization problem which reflects the presence of conflicts between different sources.