Contents

Overview

The BayesiaLab WebSimulator is a new solution which allows you to share interactive models with your audience, without having to install any software on their computer. 

The BayesiaLab WebSimulator uses your .xbl file to create a web page that is specific to your network. You can designate which nodes are accessible as Input Nodes and Output Nodes. Once the network is published, the BayesiaLab WebSimulator utilizes the Bayesia Engine API to perform inference in your network. This allows anyone to enter observations/evidence for the nodes in your network via the web. The BayesiaLab Simulator's web interface is responsive, thus allowing even tablet or smartphone users to work dynamically with your published network. Whenever evidence is set, the user will immediately see updated probability distributions of the Output Nodes.

Example

Here is a Bayesian network that shows links between numerous vehicle components that are implicated with regard to engine start. Individual components can either be causes or symptoms related to engine start.

The BayesiaLab network file (in XBL format) contains the following:

  • The Bayesian network that will be used for probabilistic inference.
  • The definition of Input Nodes and Output Nodes, along with the type of Monitor to use (gauge, slider, text).
  • Images to be associated with Monitors.
  • Comments describing the network model.

Please note that models can be published publicly or privately. Public means that anyone can potentially access your simulator and experiment with it. Private means that you can restrict access to designated users who require a model-specific ID and password.

You can publish a number of public models with your BayesiaLab license free of charge. To publish private and password-protected networks, you need to purchase an add-on license, which is available for annual rental in quantities of 5, 10, 25, 50, or 100 concurrent models.