Search Results

Bayesian networks : with examples in R

Author

Scutari, Marco, author.

Title

Bayesian networks : with examples in R / Marco Scutari, UCL Genetics Institute (UGI), Jean-Baptiste Denis, Unité de Recherche Mathématiques et Informatique Appliquées, INRA.

Format

Book

Published

Boca Raton ; London ; New York : CRC Press : Taylor & Francis Group, [2014] ©2015

Description

xv, 225 pages : illustrations ; 25 cm.

Other contributors

Denis, Jean-Baptiste, 1949- author.

Uniform series

Texts in statistical science.

Notes

"A Chapman & Hall book." Includes bibliographical references (pages 215-221) and index.

Contents

• Machine generated contents note: 1.The Discrete Case: Multinomial Bayesian Networks
• 1.1.Introductory Example: Train Use Survey
• 1.2.Graphical Representation
• 1.3.Probabilistic Representation
• 1.4.Estimating the Parameters: Conditional Probability Tables
• 1.5.Learning the DAG Structure: Tests and Scores
• 1.5.1.Conditional Independence Tests
• 1.5.2.Network Scores
• 1.6.Using Discrete BNs
• 1.6.1.Using the DAG Structure
• 1.6.2.Using the Conditional Probability Tables
• 1.6.2.1.Exact Inference
• 1.6.2.2.Approximate Inference
• 1.7.Plotting BNs
• 1.7.1.Plotting DAGs
• 1.7.2.Plotting Conditional Probability Distributions
• 1.8.Further Reading
• 2.The Continuous Case: Gaussian Bayesian Networks
• 2.1.Introductory Example: Crop Analysis
• 2.2.Graphical Representation
• 2.3.Probabilistic Representation
• 2.4.Estimating the Parameters: Correlation Coefficients
• 2.5.Learning the DAG Structure: Tests and Scores
• 2.5.1.Conditional Independence Tests

• Contents note continued: 2.5.2.Network Scores
• 2.6.Using Gaussian Bayesian Networks
• 2.6.1.Exact Inference
• 2.6.2.Approximate Inference
• 2.7.Plotting Gaussian Bayesian Networks
• 2.7.1.Plotting DAGs
• 2.7.2.Plotting Conditional Probability Distributions
• 2.8.More Properties
• 2.9.Further Reading
• 3.More Complex Cases: Hybrid Bayesian Networks
• 3.1.Introductory Example: Reinforcing Steel Rods
• 3.1.1.Mixing Discrete and Continuous Variables
• 3.1.2.Discretising Continuous Variables
• 3.1.3.Using Different Probability Distributions
• 3.2.Pest Example with JAGS
• 3.2.1.Modelling
• 3.2.2.Exploring
• 3.3.About BUGS
• 3.4.Further Reading
• 4.Theory and Algorithms for Bayesian Networks
• 4.1.Conditional Independence and Graphical Separation
• 4.2.Bayesian Networks
• 4.3.Markov Blankets
• 4.4.Moral Graphs
• 4.5.Bayesian Network Learning
• 4.5.1.Structure Learning
• 4.5.1.1.Constraint-based Algorithms
• 4.5.1.2.Score-based Algorithms

• Contents note continued: 4.5.1.3.Hybrid Algorithms
• 4.5.2.Parameter Learning
• 4.6.Bayesian Network Inference
• 4.6.1.Probabilistic Reasoning and Evidence
• 4.6.2.Algorithms for Belief Updating
• 4.7.Causal Bayesian Networks
• 4.8.Further Reading
• 5.Software for Bayesian Networks
• 5.1.An Overview of R Packages
• 5.1.1.The deal Package
• 5.1.2.The catnet Package
• 5.1.3.The pcalg Package
• 5.2.BUGS Software Packages
• 5.2.1.Probability Distributions
• 5.2.2.Complex Dependencies
• 5.2.3.Inference Based on MCMC Sampling
• 5.3.Other Software Packages
• 5.3.1.BayesiaLab
• 5.3.2.Hugin
• 5.3.3.GeNIe
• 6.Real-World Applications of Bayesian Networks
• 6.1.Learning Protein-Signalling Networks
• 6.1.1.A Gaussian Bayesian Network
• 6.1.2.Discretising Gene Expressions
• 6.1.3.Model Averaging
• 6.1.4.Choosing the Significance Threshold
• 6.1.5.Handling Interventional Data
• 6.1.6.Querying the Network
• 6.2.Predicting the Body Composition

• Contents note continued: 6.2.1.Aim of the Study
• 6.2.2.Designing the Predictive Approach
• 6.2.2.1.Assessing the Quality of a Predictor
• 6.2.2.2.The Saturated BN
• 6.2.2.3.Convenient BNs
• 6.2.3.Looking for Candidate BNs
• 6.3.Further Reading
• A.Graph Theory
• A.1.Graphs, Nodes and Arcs
• A.2.The Structure of a Graph
• A.3.Further Reading
• B.Probability Distributions
• B.1.General Features
• B.2.Marginal and Conditional Distributions
• B.3.Discrete Distributions
• B.3.1.Binomial Distribution
• B.3.2.Multinomial Distribution
• B.3.3.Other Common Distributions
• B.3.3.1.Bernoulli Distribution
• B.3.3.2.Poisson Distribution
• B.4.Continuous Distributions
• B.4.1.Normal Distribution
• B.4.2.Multivariate Normal Distribution
• B.4.3.Other Common Distributions
• B.4.3.1.Chi-square Distribution
• B.4.3.2.Student's t Distribution
• B.4.3.3.Beta Distribution
• B.4.3.4.Dirichlet Distribution
• B.5.Conjugate Distributions
• B.6.Further Reading
• C. A note about Bayesian networks
• C 1. Bayesian Networks and Bayesian statistics.

Summary

"Bayesian Networks: With Examples in R introduces Bayesian networks using a hands-on approach. Simple yet meaningful examples in R illustrate each step of the modeling process. The examples start from the simplest notions and gradually increase in complexity. The authors also distinguish the probabilistic models from their estimation with data sets. The first three chapters explain the whole process of Bayesian network modeling, from structure learning to parameter learning to inference. These chapters cover discrete Bayesian, Gaussian Bayesian, and hybrid networks, including arbitrary random variables. The book then gives a concise but rigorous treatment of the fundamentals of Bayesian networks and offers an introduction to causal Bayesian networks. It also presents an overview of R and other software packages appropriate for Bayesian networks. The final chapter evaluates two real-world examples: a landmark causal protein signaling network paper and graphical modeling approaches for predicting the composition of different body parts. Suitable for graduate students and non-statisticians, this text provides an introductory overview of Bayesian networks. It gives readers a clear, practical understanding of the general approach and steps involved."--Back cover.

Subject headings

Bayesian statistical decision theory. Bayesian statistical decision theory--Data processing. R (Computer program language)

ISBN

9781482225587 (Hardback) 1482225581 (Hardback) 9781482225617 9781482225594 9781482225600