Amid the escalating concerns of climate change and the mounting research papers on carbon neutrality and waste-to-energy solutions, comprehending crucial knowledge and technological trends in the chemical and energy sectors has become challenging. This study presents a novel approach combining large language models (LLM) and knowledge graphs (KG) to facilitate AI-supported knowledge retrieval. This work establishes a knowledge graph in the biochemical industry with 6,461 nodes and 8,969 relationships, emphasizing material and energy flow integration with the autonomous AI workflow. The graph’s node attributes and relationships are analyzed using cosine similarities, with the capability to trace back to original literature through DOIs. This method not only underscores the relevance of node pairs in the graph but also links their similarities to the physical and chemical properties of materials. To sum up, this work provides an AI-enhanced tool that enables researchers and decision-makers to quickly build a knowledge base, learn about trends, and gain insights into specific fields.
Keywords Natural Language Processing, Knowledge Graph, Interpretability, Material Flow Integration, Sustainable Development