On the Interaction Between Inverse Features and Path-Functional Dependencies in Description Logics
Abstract
We investigate how inverse features can be added to a boolean complete description logic with path-functional dependencies in ways that avoid undecidability of the associated logical implication problem. In particular, we present two conditions that ensure the problem remains EXPTIME-complete. The first is syntactic in nature and limits the form that dependencies may have in argument terminologies. The second is a coherence condition on terminologies that is sufficiently weak to allow the transfer of relational and emerging object-oriented normalization techniques.
Cite
Text
Toman and Weddell. "On the Interaction Between Inverse Features and Path-Functional Dependencies in Description Logics." International Joint Conference on Artificial Intelligence, 2005.Markdown
[Toman and Weddell. "On the Interaction Between Inverse Features and Path-Functional Dependencies in Description Logics." International Joint Conference on Artificial Intelligence, 2005.](https://mlanthology.org/ijcai/2005/toman2005ijcai-interaction/)BibTeX
@inproceedings{toman2005ijcai-interaction,
title = {{On the Interaction Between Inverse Features and Path-Functional Dependencies in Description Logics}},
author = {Toman, David and Weddell, Grant E.},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2005},
pages = {603-608},
url = {https://mlanthology.org/ijcai/2005/toman2005ijcai-interaction/}
}