dice-group / Ontolearn

Ontolearn is an open-source software library for explainable structured machine learning in Python. It learns OWL class expressions from positive and negative examples.
https://ontolearn-docs-dice-group.netlify.app/index.html
MIT License
41 stars 9 forks source link

Refinement operation in different DLs #411

Open Demirrr opened 5 months ago

Demirrr commented 5 months ago

We need to implement subsumption based refinement operators (e.g. CELOE's) in different description logics (e.g. EL, ALC, SROIQ). Ideally, we should be able to use these refinement operators across models without an effort.

By this, we could more objectivly quantify the benefits of choosen DL in learning OWL class expressions. For instance, although EL is a subclass of ALC, to tackle a given learning problem (i.e., geneneralize well) we may not need to learn in ALC. Perhaps, EL would suffice. By this, we can compare the runtimes of using the same learner in different DLs.