TRIQ: Streamlining Semantic RDM Through Intuitive User Interfaces with Ontology Driven Decision Tree Support

Research data management (RDM) in catalysis and engineering disciplines faces increasing challenges due to metadata heterogeneity, volume, and complexity. The adoption of semantic technologies offers great potential for structured metadata representation and knowledge integration. However, current RDF-based systems often lack user-friendly interfaces, efficient query mechanisms, and scalable workflow integration. To address these challenges, we present TRIQ [1], a novel metadata management platform designed to streamline the creation, storage, and retrieval of RDF-based research data. TRIQ offers a scalable, user-centric solution that enables researchers to manage semantic metadata without requiring deep expertise in RDF, SPARQL or ontology engineering. The platform integrates intuitive query builders and seamless interoperability with existing research infrastructures. TRIQ can be used either as a standalone web application, or function as an external tool for a central Repo4Cat data repository [2], providing extended functionality and automated metadata extraction. In this setup, the platform automatically extracts existing user-provided metadata, ensuring consistency across systems while minimizing duplication of effort. The system also adheres to NFDI's AAI [3] standards, enabling secure, standardized authentication and authorization. The most useful features of TRIQ are listed below: • Metadata fields are not hard-coded. Instead,they are dynamically generated based on the concepts and relations from the Metadata4Cat ontology [4], which ensures semantic correctness and future-proofing, as well as prevents users from entering inconsistent data. • Dynamic questionnaires are generated and progressively extended through ontology-based reasoning over the data provided by users. • Metadata is stored natively as RDF triples in a triple store. • Support for exporting data in standard RDF formats. • As the questionnaire progresses, the generated metadata becomes instantly accessible for browsing which in conjunction with natural language querying greatly lowers the barrier for non-technical users to access linked data. The TRIQ platform is built on a modular architecture. Its key components include a user-friendly frontend for metadata form creation and visualization, a real-time WebSocket bridge via Node.js, and a core Java questionnaire engine for dynamic field generation and reasoning-based data processing. External modules connect seamlessly with services such as PID4Cat [5] and Repo4Cat. This architecture provides a flexible foundation for extending functionality and integrating with external RDM services. The TRIQ platform employs TDB2, Apache Jena's native RDF store, as its semantic backend, seamlessly integrated with Apache Jena Fuseki — an open-source SPARQL server that facilitates efficient querying and updating of RDF data, provides ontology support, access control, and strong standards compliance. Apache Jena library was selected for its flexibility, stability, federated query capabilities, ease of deployment, active community, and high level of customization. Together, TRIQ and Apache Jena's native RDF store enable efficient metadata generation, structured storage, and semantic querying. Currently, TRIQ continues to evolve through enhanced metadata enrichment, advanced semantic search and filtering, and stronger integration with cross-domain ontologies and linked open data ecosystems. By embedding semantic technologies into user-friendly workflows, TRIQ positions itself as a key enabler of sustainable, FAIR-compliant RDM in catalysis and related fields.