TRIPLE ONTOLOGY: A REUSABLE FORMALISATION OF SCIENTIFIC ENTITIES IN THE SOCIAL SCIENCES AND HUMANITIES DOMAIN

The TRIPLE project, launched in October 2019 and coordinated by the French National Center for Scientific Research (CNRS), involved 22 partners from 15 European countries. Its primary aim was to develop the GoTriple discovery platform, a multilingual access point for discovering and reusing research artefacts in the social sciences and humanities (SSH). This thesis presented a comprehensive ontology designed to formalise the TRIPLE data model using semantic technologies. The ontology addressed the challenge of managing heterogeneous data aggregated by the Core Pipeline, which integrated research artefacts from diverse external sources with varying structures and formats.This newly developed ontology ensured a robust semantic representation of research artefacts, surpassing the limitations of initial alignments with Schema.org and SIOC standards. It was aligned with the main data aggregators, ensuring adherence to state-of-the-art practices in the field. Developed in collaboration with domain experts, the ontology was crafted to achieve several key objectives: formalising the data model with semantic standards, defining controlled vocabularies, establishing connections with external entities, ensuring resource reusability, and maintaining detailed documentation for transparency and extensibility. The ontology development followed a structured methodology, beginning with a preliminary analysis and employing the Simplified Agile Methodology for Ontology Development (SAMOD). This approach ensured a flexible, iterative development process with comprehensive testing and documentation, guaranteeing the ontology's accuracy and adaptability. Consequently, the ontology enhanced the semantic representation of research artefacts, promoted interoperability, and facilitated collaboration and knowledge reuse across the SSH domain.