Towards an automatic SPARQL query generation from ontology competency questions

Abstract

Ontologies are omnipresent and are actively used as explicit, formal, and consensual knowledge representations. Hence, ensuring ontological quality is considered a crucial issue. One of the well-known domain ontologies validation approaches proposes a set of competency questions, covering the domain represented by the ontology, translating these questions into SPARQL queries, running them on the ontology, and comparing the obtained results with the expected ones. This paper presents an approach to translate natural language competency questions into SPARQL queries to assist domain experts in testing ontologies and users considering exploring ontologies for further exploitation. The implementation and experimentation of the proposed approach are described. The resulting tool proposes a fully automatic and semi-automatic execution, enabling end-users to parameter the approach phases.

KEYWORDS:

• Question Answering System
• Natural Language Processing
• ontology validation
• SPARQL
• Entity Mapping
• competency question

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 https://www.dbpedia.org/resources/ontology/

2 https://github.com/yago-naga/yago4

3  https://softwareontology.wordpress.com/2011/04/01/user-sourced-competency-questions-for-software.

Additional information

Notes on contributors

Karim Benhocine

Karim Benhocine received his Computer Science Engineering diploma and his Research Master in Computer Science in 2020 from ESI (Ecole Nationale Supérieure d’Informatique), Algiers, Algeria. He is currently a Data Scientist working in freelance

Adel Hansali

Adel Hansali received his Computer Science Engineering diploma and his Research Master in Computer Science in 2020 from ESI (Ecole Nationale Supérieure d’Informatique), Algiers, Algeria. He is currently a certified SAP Consultant working in freelance while looking for Ph.D. offers in artificial intelligence and information technologies fields.

Leila Zemmouchi-Ghomari

Leila Zemmouchi-Ghomari received her Ph.D. in Computer Science from ESI (Ecole Nationale Supérieure en Informatique), Algiers, in January 2014. Her research interests focus on ontology engineering, knowledge engineering, semantic web, and linked data. She is currently an associate professor at Industrial Engineering and Maintenance Department at ENST: Ecole Nationale Supérieure de Technologie, Algiers, Algeria.

Abdessamed Reda Ghomari

Abdessamad Reda Ghomari received his Ph.D. in Computer Science from ESI (Ecole Nationale Supérieure d’Informatique), Algiers, in 2008. He is currently a Full Professor at ESI, Algiers, Algeria. From 2001 to 2021, he is Information System Management Team Head at LMCS (the Laboratory of Systems Design Methodologies). His research interests focus on knowledge engineering, Crisis Informatics, Open Government Data, Organization 2.0 & Data Governance. He has supervised the activities of partnerships with enterprises and continuing training (2011-2018).