Inference Rules are used for discovering new relationships using automatic procedures. This module utilizes learning algorithms and site data and / or metadata to generate new knowledge.
It currently includes a Markov Chain library, which should be moved to DAIL; a hard coded action for querying views data, which should be replaced with Views Rules; and a custom coded 'Set field value with Markov Chain', which should be removed in favor of 'Set a data value' and other generic actions.
This code was originally intended for testing some theories on concept reflection and ontology alignment. It was, however, only tested in a highly scaled back use case of generating recipes based on common ingredients in cuisines from various cultures. Other uses might include, generating site content from existing articles, or even more advanced concept reasoning. A related effort can be found in #326176: Using Rules with RDF
. For more information on the intended use cases, see:
* Programming done with Christopher Leppanen.