Skip to Main content Skip to Navigation

Chase Variants & Boundedness

Stathis Delivorias 1
1 GRAPHIK - Graphs for Inferences on Knowledge
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : The chase is a family of algorithms designed to infer data with the use of ontological knowledge, which encoded in existential rules, a sub-language of first-order logic. A considerable literature has been devoted to its analysis, approaching it from a variety of presupposed terminological and notational background. We define a unifying framework for the specification and study of chase algorithms. We utilize it to compare and clarify the properties that discern the different variants of the chase. We particularly focus on studying whether there is a bound in the maximum length of a chain of interdependent rule applications (where interdependency means that the output of a rule application is contributing to triggering the next rule application). This is the problem ofboundedness, or k-boundedness, when the bound k is given. By investigating a number of intermediate properties, we find that k-boundedness is decidable for several chase variants. In addition to other secondary results, we define two new chase variants with the aim of reducing redundant rule applications without heavily increasing the computation cost.
Complete list of metadatas

Cited literature [66 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Wednesday, February 12, 2020 - 3:04:16 PM
Last modification on : Wednesday, October 14, 2020 - 4:25:28 AM
Long-term archiving on: : Wednesday, May 13, 2020 - 4:32:20 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02408597, version 2



Stathis Delivorias. Chase Variants & Boundedness. Data Structures and Algorithms [cs.DS]. Université Montpellier, 2019. English. ⟨NNT : 2019MONTS039⟩. ⟨tel-02408597v2⟩



Record views


Files downloads