DEV Community

Helio
Helio

Posted on

I Wrote a Knowledge Graph Render by LangChain

Hello and good morning.

I am a student from Chain. Recently, our teacher let us to write a KnowledgeGraph(as KG) render and save the graph to a graph database like neo4j.

Yeah, I know LangChain already provide the function to generate the graph data, but I still want to build my own version of KG. I tried many models like phi-3-mini, llama3-8b, but none of them work very well, so I changed my mind to use GLM-4 powered by ZhiPuAI, and it works very well. Here is the showcase: A very large image.

KG

Although the showcase shows that KG render works very well, but I have some problem to fix:

  • I save graph to neo4j, but for the same label nodes, they are all connected to a same label nodes, although they have different type. Just like the word Key, Key could be related to many different fields like keyboard or database.

  • If I enable the update in args, render image will slow down the speed entirely, If there is a better way to show real time graph?

Image of Timescale

🚀 pgai Vectorizer: SQLAlchemy and LiteLLM Make Vector Search Simple

We built pgai Vectorizer to simplify embedding management for AI applications—without needing a separate database or complex infrastructure. Since launch, developers have created over 3,000 vectorizers on Timescale Cloud, with many more self-hosted.

Read full post →

Top comments (0)

Image of Docusign

🛠️ Bring your solution into Docusign. Reach over 1.6M customers.

Docusign is now extensible. Overcome challenges with disconnected products and inaccessible data by bringing your solutions into Docusign and publishing to 1.6M customers in the App Center.

Learn more