Metric Temporal Logic for Ontology-Based Data Access over Log Data

TL;DR

This paper introduces HornMTL, a metric temporal logic, and its datalog extension datalogMTL, demonstrating their application in ontology-based data access over meteorological log data, with decidability results for certain query classes.

Contribution

It presents a new metric temporal logic and its datalog extension, establishing decidability for ontology-mediated queries in a practical non-recursive fragment.

Findings

Decidability of answering ontology-mediated queries for non-recursive datalogMTL.

Application of HornMTL and datalogMTL in meteorological data analysis.

Potential use-cases in engine and device log data analysis.

Abstract

We present a new metric temporal logic HornMTL over dense time and its datalog extension datalogMTL. The use of datalogMTL is demonstrated in the context of ontology-based data access over meteorological data. We show decidability of answering ontology-mediated queries for a practically relevant non-recursive fragment of datalogMTL. Finally, we discuss directions of the future work, including the potential use-cases in analyzing log data of engines and devices.