The graph maker library discussed in the content improves upon previous approaches by finding a balance between rigour and ease, structure and lack of it. It performs better on various challenges compared to previous methods. Unlike the previous approach where the LLM discovers the ontology by itself, the graph maker enforces the use of a user-defined ontology. The process involves defining the ontology, splitting the text into chunks, converting the chunks into documents, running the graph maker, and saving the results to Neo4j for visualization. The library uses prompts to generate consistent JSONs and can handle parsing failures. The final graph consists of edges represented as pydantic models. The graph can be visualized using Neo4j’s Bloom application. Additionally, the graph maker can track the evolution of relationships between characters in a book by adding a temporal dimension to the graph through the ‘order’ attribute in the edge model. This allows for a visual representation of how relationships evolve over time.
Source link
Source link: https://towardsdatascience.com/text-to-knowledge-graph-made-easy-with-graph-maker-f3f890c0dbe8
#GraphMaker simplifies converting text into knowledge graph effortlessly. #KnowledgeGraph
![Text to Knowledge Graph Made Easy with Graph Maker | by Rahul Nayak | May, 2024](https://i0.wp.com/webappia.com/wp-content/uploads/2024/05/15VFKT7FPjzE_MkQMq3NXfA.png?fit=758%2C313&quality=80&ssl=1)
GIPHY App Key not set. Please check settings