inst/extdata/pavement_example.md

In bearloga/shinyBN: Interactive Visualization for Bayesian Networks

Pavement Model

The included example model pavement_model is a Bayesian network created with the bnlearn package. It's a directed acyclic graph (DAG) which describes the probabilistic relationships between Rain (Yes/No), Sprinkler (On/Off), and Pavement (Wet/Dry).

The joint probability function is:

$$ \text{Pr}(\mathbf{P}, \mathbf{S}, \mathbf{R}) = \text{Pr}(\mathbf{P}|\mathbf{S},\mathbf{R}) \text{Pr}(\mathbf{S}|\mathbf{R}) \text{Pr}(\mathbf{R}) $$

This Bayesian network can be used to answer probabilistic queries about the variables even when only a subset of variables are known. The observed nodes are evidence and approximate inference algorithms -- together with Bayes' rule -- propagate this evidence throughout the network to update the probabilities in the unobserved nodes.

Further Reading

• Fenton, N., & Neil, M. (2018). Risk Assessment and Decision Analysis with Bayesian Networks, Second Edition. CRC Press.
• Scutari, M., & Denis, J. (2014). Bayesian Networks: With Examples in R. CRC Press.
• Nagarajan, R., Scutari, M., & Lè̀bre, S. (2013). Bayesian Networks in R. Springer Science & Business Media. http://doi.org/10.1007/978-1-4614-6446-4
• Kjærulff, U. B., & Madsen, A. L. (2007). Bayesian Networks and Influence Diagrams: A Guide to Construction and Analysis. Springer Science & Business Media. http://doi.org/10.1007/978-0-387-74101-7
• Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Springer. (PDF)

bearloga/shinyBN documentation built on May 25, 2019, 4:01 a.m.