WebSimulator Quick Start Guide
Deploying a Bayesian Network Model via the Web
For this Quick Start Demo, we have prepared a simple Bayesian network model, which captures the relationships between house characteristics and property value. The objective is to compute the expected value of the target node Sale Price as a function of observed property attributes. Once this model is published via the WebSimulator, end users can set observations interactively via the web interface and instantly obtain the predicted values.
The network was learned from a dataset that contains the sales prices of residential properties in Ames, Iowa, from 2006 to 2010. This dataset was first used by De Cock (2011) as an educational tool for statistics students.
This Quick Start Guide shows you step-by-step how you can take this Bayesian network model and instantly publish it via the BayesiaLab WebSimulator. Further details about the WebSimulator and its administration can be found in the Bayesia Library:
Anyone with a BayesiaLab Professional license can use the WebSimulator to publish models free of charge to the general public. Additional subscription fees apply for publishing a model with a private and secure account on the WebSimulator server. Please contact us at firstname.lastname@example.org for details.
De Cock, D. (2011). Ames, Iowa: Alternative to the Boston Housing Data as an End of Semester Regression Project. Journal of Statistics Education. www.amstat.org/publications/jse/v19n3/decock.pdf