Mapping the Invisible Graph Powered Archaeology with Neo4j, LiDAR, and LLMs
summary
A working system called Archaios combines LiDAR-derived layer extraction, Azure Document Intelligence layout OCR, semantic chunking and embedding, and a Neo4j vector index to store embeddings plus domain nodes; Autogen agents coordinate terrain segmentation and image analysis via GPT-4 Vision, flagging candidates as hypotheses in the graph and sending notifications for archaeologist confirmation. The pipeline links lexical and domain graphs for multihop retrieval, and a live demo surfaces multiple candidate sites for human validation.
AI-written summary · about
chapters (11)
- 0:00Introduction and context
- 1:14Defining the problem space
- 4:25Agenda overview
- 5:22Understanding RAG approaches
- 9:22Introduction to graph RAG
- 11:14Lexical graph ingestion
- 16:40Chunking and enrichment
- 21:52Lexical graph construction
- 23:48Building the domain graph
- 25:18Bridging graphs and retrieval
- 28:07Live demo and conclusion
related talks
30mNeurosymbolic Architecture in Production with Neo4j, GDS, Node2Vec & LLM Grounding
08mHarness Graph Power With Agentic AI: An Insights Discovery Use Case
28mGrounded Intelligence Boosting LLMs with Real-Time Graph Context
23mConnecting the Dots With Context Graphs
24mUnlocking AI's Hidden Connections With Graphs
36mGraphRAG Explained 🔥 Building Smarter LLMs
01h 00mGraph Mastery to AI
01h 17mGraph Database and Retrieval Augmented Generation