This example illustrates the use of a Bayesian Belief Network (BBN) for predicting yield at the farm level. Assume we have a farmer who is growing crops that would like to predict the yield of this year's harvest based on some specific knowledge of the farm. In this example, the farmer may grow rape, barley, wheat and oat.
A BBN will be used to compute a probability distribution over the yield (in hkg). To support the reasoning about the expected yield, we assume the farmer knows the grain type produced last year and the soil type of the field.
Here are some HUGIN widgets for interacting with the model shown on the right (click on the probability bar to instantiate a node or remove evidence):
The expected yield is with variance .
In the example the field 7-0 had common wild oat last year. Notice the spillover effect on neighboring fields.
As an example, the expected yield is 82.20 hkg for a wheat field on clay with grain type last year being wheat.
Useful references for those interested in BBN include:
Kjærulff, U. B. and Madsen, A. L. (2013) Bayesian Networks and Influence Diagrams: A Guide to Construction and Analysis. Springer, Second Edition.
Martin Karlsen, HUGIN EXPERT A/S, mk at hugin dot com
Anders L Madsen, HUGIN EXPERT A/S, anders at hugin dot com