Projects
Making Smart Materials Smarter: How Semantic Web Development Enhances livMatS
My project aims to develop a semantic web and ontologies for describing and structuring material properties. Unlike the traditional web, the Semantic Web organizes data in a way that allows machines to interpret and use it. This organization is achieved through ontologies. Ontologies are structured frameworks that define the concepts within a domain and the relationships between them. Think of an ontology as a detailed map that categorizes and connects different pieces of knowledge. In the context of LivMatS, ontologies might include concepts like "self-repair," "energy harvesting," and "adaptive behavior," along with explanations of how these concepts interrelate.
By using ontologies, we can create a well-structured and interconnected knowledge base that makes it easier for machines to process and analyze data. This capability is crucial for developing the advanced, intelligent materials systems envisioned by livMatS.
Project outcomes
I designed and implemented AI-assisted tools that automatically generate semantic metadata, allowing researchers to organize experimental and simulation data more efficiently. I developed and validated domain-specific ontologies in materials science, which improved data consistency and made it easier to share and reuse data across multiple research projects. Large language model (LLM) pipelines were integrated into existing research workflows, significantly reducing the time required for manual annotation and metadata creation while maintaining a high level of data quality. To ensure semantic correctness and compliance with predefined standards, I implemented validation mechanisms based on RDF, OWL, and SHACL. In addition, I deployed scalable web applications using Django/FastAPI and PostgreSQL, making the developed tools accessible to interdisciplinary research teams within the project partners.