Ceres

Ce
Ceres

Creator:Sjaak van Werkhoven
Function:Rule based reasoning: implementation of Moku and Nabby results
Input Type:xml, txt
Input from Agent:Moku, Nabby
Output Type:Classification advice
Output to Agent:Advice, Juno
Short Description:

The Ceres agent allows us to implement classification in the form
of If -> Then rules.

Rules are based on branches from Moku classification trees, combinations of Nabby settings, but may also represent Health Care guidelines or expert decision rules (based on literature or local practice). Rules and selection of sets of rules may be altered on the fly.

The performance of these rules can be tested with testcases where the expected outcome can be compared with the result of the used rules. The deduction of the advice can be shown by reasoning traces in text or graphical form.





© 2015 Alan Turing Institute Almere