Knowledge Graphs with Obsidian

Summary

Llama Index is a python package that can be used to create Knowledge graphs (KGs). There exists a method to integrate with Obsidian. This is a tutorial on how to set this up a RAG system with obsidian. Visualisation is done using the pyvis library.
Requires API to LLM.

Important

• RAG improves LLM performance by utilizing external databases.
• Llama Index facilitates the transformation of Obsidian notes into a structured Knowledge Graph.
• The tutorial includes steps for setup, dependencies, and visualization.

Attention

• LLMs can still produce errors known as “hallucinations.”
• Requires familiarity with Python and Jupyter Notebook.

Code

Example

The tutorial provides a code snippet for querying the Knowledge Graph about the assumptions of the Black-Scholes model, yielding detailed contextually relevant insights.

Follow

• How can RAG be applied to other data sources beyond Obsidian?
• What specific challenges might arise when integrating large datasets into a Knowledge Graph?
• Can you also set up with PDF’s within Obsidian?

Link

https://medium.com/@haiyangli_38602/make-knowledge-graph-rag-with-llamaindex-from-own-obsidian-notes-b20a350fa354 Knowledge Graph