Automated Construction of a Knowledge Graph of Nuclear Fusion Energy for Effective Elicitation and Retrieval of Information

In this document, we discuss a multistep approach to automated construction of a knowledge graph (KG), for structuring and representing domain-specific knowledge from large document corpora. We apply our method to build the first KG of nuclear fusion energy, a highly specialized field characterized by vast scope and heterogeneity. This is an ideal benchmark to test the key features of our pipeline, including automatic named entity recognition (NER) and entity resolution. We show how pretrained large language models (LLMs) can be used to address these challenges and we evaluate their performance against Zipf’s law, which characterizes human natural language. In addition, we develop a knowledge-graph retrieval-augmented generation (RAG) system that uses multiple prompts with LLMs to provide contextually relevant answers to natural-language queries, including complex multihop questions requiring reasoning across interconnected entities.