Building Ontologies and Knowledge Graphs traditionally relies on human expertise. The paper From human experts to machines explores how open-source LLMs can help automate aspects of this process. The proposed pipeline includes competency questions formulation, TBox ontology development, construction, and evaluation—all facilitated by LLMs. A benchmark study with a focus on Alzheimer’s Disease showcases how LLMs, grounded in a caregiving knowledge base using the RAG framework, can create more reliable and accessible tools for caregivers.

• Human Involvement Minimization: Coines a process that significantly reduces the need for human expertise in KG construction.
• LLM-Powered Pipeline: Designs a pipeline that incorporates LLMs from conceptualization to evaluation.
• Domain-Specific Application: Demonstrates potential in caregiving, one of the many fields that could benefit from this automation.

This research could mark a milestone in employing AI for structured knowledge creation, simplifying complex tasks and possibly leading to a knowledge revolution across multifarious disciplines. Glimpse the future of knowledge organization powered by AI.