Over ons

Consortium

Team

Vlaams Beleidsplan AI

Maak kennis met onze Alumni

Contact

<h5>Description:</h5>
<p>The Decision Model and Notation (DMN) standard is a user-friendly notation standard for decision logic. It is easily readably by everyone involved in &ldquo;decision management&rdquo;, such as business people, IT experts and business analysts. However, the current usage of DMN &ndash; in which every model only has one way of executing it &ndash; does not make use of its full potential.</p>
<p>The DMN-IDPy API, based on the <a href="dmn-idp">DMN-IDP</a> interface, combines a DMN modeler with the IDP knowledge based system as an execution engine. This way, more flexible usages for a DMN model are &ldquo;unlocked&rdquo;. Indeed, usage of a DMN model is no longer limited to only the &ldquo;bottom-to-top" approach. Indeed, among others, the API also allows &ldquo;multidirectional reasoning&rdquo;, in which any variable of a DMN model can be calculated from any other variables, thus removing the limitation of the &ldquo;bottom-to-top" execution. A list of all features:</p>
<ul>
<li>Load in DMN models via XML</li>
<li>Execute models using IDP, without having to know or learn it</li>
<li>&ldquo;Standard&rdquo; bottom-to-top execution approach</li>
<li>Multidirectional reasoning</li>
<li>Reasoning with incomplete information</li>
<li>Relevance (I.e., which variables should be known to calculate another one)</li>
<li>Querying the already &ldquo;known variables&rdquo;</li>
<li>Querying the variable type, and list of possible values</li>
<li>Optimization</li>
</ul>
<p>The overarching goal of DMN-IDPy is to allow a user-friendly representation of knowledge in the form of DMN to build knowledge-based AI. Consider e.g. a chat bot, that has its &ldquo;brain&rdquo; encoded as a DMN model, or the behavior of a cobot modeled in simple DMN.</p>

<h5>More Information:</h5>
<ul>
<li><strong>Grand Challenge/Workpackage(s)</strong>: <a href="/nl/onderzoek/onderzoeksuitdagingen-2019-2023/ai-driven-data-science"[entry:58:/onderzoek/onderzoeksutidagingen-2019-2023/ai-driven-data-science]data-entry-id="58" target="_blank" rel="noopener">GC1</a> - WP2</li>
<li><strong>Responsible Research Lead</strong>: <a href="mailto:Joost.vennekens@kuleuven.be" target="_blank" rel="noopener">Joost Vennekens</a></li>
<li><strong>Authors</strong>: <a href="mailto:s.vandevelde@kuleuven.be" target="_blank" rel="noopener">Simon Vandevelde</a></li>
<li><strong>Programming Language</strong>: Python</li>
<li><a href="https://cdmn.readthedocs.io/en/latest/DMN_guide.html" target="_blank" rel="noopener">Link to the website</a></li>
<li><a href="https://gitlab.com/EAVISE/cdmn/cdmn-solver" target="_blank" rel="noopener">Link to the repository</a></li>
<li><strong>Categories</strong>: Symbolic AI</li>
</ul>

<h5>Publications:</h5>
<ul>
<li>
<p>S. Vandevelde, V. Etikala, J. Vanthienen, J. Vennekens, "Leveraging the Power of IDP with the Flexibility of DMN: a Multifunctional API", in Proceedings of RuleML+RR21, 2021</p>
</li>
</ul>