Accelerating (Biomedical) Knowledge Graph Construction with LLMs
What does the day in the life of a medical specialist who encounters a patient with an unclear diagnosis look like? Its combing through tens or maybe hundreds of scientific papers to find a gene, cell therapy or something else that may be the key to saving their patient’s life. As you can imagine, this can be a lengthy and time-consuming process. But what if there was a tool this specialist could use to get this information through simple queries, cutting down the amount of time it takes to find the needed information?
Here enter knowledge graphs. In this blogpost, I’m going to walk you through how to build a knowledge graph using Large Language Models (LLMs) to empower biomedical research. This approach can be exploited for other use cases that require organising information from diverse unstructured sources into a structured format — and even for your graphRAG applications.
What are knowledge graphs anyway?
Knowledge graphs simply represent data as an interconnected network of entities(nodes) and relationships.
Let’s take Acute Myeloid Leukemia, a type of cancer of the blood and bone marrow that affects the production of certain blood cells in humans. Representing this as a graph, our 3 entities, “disease”, “cell” and “tissue” are depicted as nodes and the relationships between them depicted by the arrows.
Accelerating (Biomedical) Knowledge Graph Construction with LLMs was originally published in ML6team on Medium, where people are continuing the conversation by highlighting and responding to this story.
