Learning Bayesian Networks with the bnlearn R Package

TL;DR

The paper introduces bnlearn, an R package that offers various algorithms for learning Bayesian network structures, supporting both discrete and continuous variables, with features for parallel computing and advanced visualization.

Contribution

It provides a comprehensive R package integrating multiple structure learning algorithms, scoring methods, and visualization tools for Bayesian networks, enhancing usability and performance.

Findings

Supports both constraint-based and score-based algorithms

Enables parallel computing for faster learning

Includes advanced visualization options

Abstract

bnlearn is an R package which includes several algorithms for learning the structure of Bayesian networks with either discrete or continuous variables. Both constraint-based and score-based algorithms are implemented, and can use the functionality provided by the snow package to improve their performance via parallel computing. Several network scores and conditional independence algorithms are available for both the learning algorithms and independent use. Advanced plotting options are provided by the Rgraphviz package.