Optimizing Ontology Materialization with Equivalent Role and Inverse Role Rewriting
Xiaoyu Qin, Xiaowang Zhang, Zhiyong Feng · Companion Proceedings of the Web Conference 2020 · 2020
Materialization, as a basic approach to ontology-mediated query answering, extends an ontology by exploiting the semantic knowledge expressed in the ontology to improve query answering over data. However, those existing materialization approaches are often inefficient, especially when ontologies consist of complex roles. In this paper, we propose a new rewriting algorithm for equivalent roles and inverse roles to both improve materialization efficiency and reduce memory consumption. Moreover, we develop a role scoring algorithm to optimize role rewriting. The preliminary experiments show that our approach can significantly reduce materialization times and memory consumption.