We offer a 2 day training course in Bayesian networks, using Bayes Server™. We offer both private and public training events.


Although the Bayes Server™ APIs are cross platform, the course makes use of the Bayes Server™ user interface which is windows only.

Each delegate must bring a laptop with a suitable version of Windows installed. Instructions to install the required software will be provided to delegates before the course.

Note that Windows can be installed on Mac OS X using BootCamp or as a Virtual Machine.

Private events


If you would like us to conduct training within your organisation, please feel free to contact us with your requirements.


Contact us


Public events

Upcoming public events are listed below.


Please confirm places are still available before purchasing. You can purchase online, or send us a purchase order.


Country Date Location Venue Book online



Subscribe to training event notifcations




Course outline

A course outline is provided below. This is subject to change.

Day 1


  • What is a variable?
    • Variables - discrete & continuous
    • Discretization?
  • Probability foundations
    • Probability
    • Joint probability
    • Conditional probability
    • Marginal probability
    • Multiplication
    • Instantiation
    • Division
    • Bayes Theorem
    • Conditional independence
    • Exercise - probability
  • Bayesian networks - construction
    • Bayesian networks - Definition
    • Adding nodes & variables
    • Nodes from data
    • Adding links
    • Parameter counter
    • Links from data
    • Distributions & parameters
    • The node toolbar
    • Exercise - construction
  • Bayesian networks - inference part I
    • Adding queries
    • Query explorer
    • Evidence
    • Prediction & diagnostics
    • Converging, diverging and serial connections
    • Exercise - inference
  • Inference with data
    • What is a case?
    • Using databases and spreadsheets
    • Data explorer
    • Batch queries
    • Mesh queries
    • Exercise - data connection
    • Exercise - inference with data
  • Bayesian networks - parameter learning part I
    • Learning parameters
    • Candidate networks
    • Log likelihood
    • Exercise - parameter learning
  • Bayesian networks - structural learning
    • Structural learning algorithms
    • Exercise - structural learning
  • Analysis - part I
    • Value of information
    • Pattern Analysis
    • Automated insight
    • Link Strength
    • D-separation
  • Decision graphs
    • Understanding Decision Graphs
  • Latent variables
    • Latent variables
  • Segmentation/clustering
    • What is clustering?
    • Mixture models
    • Exercise - clustering
  • Regression (Optional)
    • What is regression?
    • A simple model - linear regression
    • Exercise - regression
  • Classification (Optional)
    • What is classification?
    • A simple model - Naive Bayes
    • Confusion matrix
    • Lift chart
    • Exercise - classification


Day 2


  • Noisy nodes
    • Noisy nodes
  • Mult-variate nodes
    • Multi-variable nodes
    • Decomposition
    • Exercise - multi variable nodes
  • Bayesian networks - inference part II
    • Joint queries
    • Log likelihood
    • Conflict
    • Retracted evidence
    • Most Probable Explanation
    • Comparison queries
    • Exercise - inference part II
    • Basic algorithm - Variable elimination
    • Inference algorithms
  • Analysis - part II
    • Impact Analysis
    • Log-likelihood Analysis
    • Exercise - Log-likelihood Analysis
    • Tree width
    • Sensitivity Analysis
    • Parameter tuning
    • In-Sample Anomalies
    • Retracted analysis
  • Modelling techniques
    • Divorcing
    • Expert disagreements
    • Exercise - modelling techniques (Optional)
  • Dynamic Bayesian networks
    • Temporal variables
    • Temporal links
    • Temporal distributions
    • Unrolling
    • Exercise - unrolling
  • Dynamic Bayesian networks - inference
    • Queries (time series)
    • Query explorer (time series queries)
    • Temporal evidence
    • Data explorer (time series data)
    • Exercise - time series inference
  • Bayesian networks - parameter learning part II
    • Filtering
    • Initialization
    • Time series distributions
    • Learning with missing data
    • Case weights
    • Exercise - parameter learning part II
  • Bayesian networks - online learning
    • What is online learning (adaptation)?
  • Bayesian networks - data sampling
    • Generating sample data
    • Charting sampled data
    • Missing data
    • Writing to a data source
    • Exercise - data sampling
  • Anomaly detection
    • What is anomaly detection?
    • Exercise - anomaly detection
  • Bayesian networks - using the Bayes Server library (API)
    • Exercise - using the Bayes Server library
  • Time series models (Optional)
    • Types of time series model
    • Prediction, filtering, smoothing and most probable sequences
    • Exercise - time series models
  • Arc Reversal (Optional)
    • Arc reversal
    • Exercise - Arc Reversal


User interface

  • Windows 10 (x86 and x64)
  • Windows Server 2016 (x86 and x64)
  • Windows 8 (x86 and x64)
  • Windows Server 2012 (x64)
  • Windows 7 (x86 and x64)
  • Windows Vista (x86 & x64) with Service Pack 2
  • Windows XP (x86) with Service Pack 3
  • Windows Server 2008 (x86 and x64) with Service Pack 2
  • Windows Server 2008 R2 (x64)
  • Windows Server 2003 (x86 & x64) with Service Pack 2
  • Windows Server 2003 R2 (x86 and x64)