Graph machine learning seminar work at MaCAD, IAAC: notebooks on graph representation, embeddings and classification.
-
Updated
Aug 4, 2026 - Jupyter Notebook
Graph machine learning seminar work at MaCAD, IAAC: notebooks on graph representation, embeddings and classification.
The Narkomfin Building as a spatial graph: centrality and community detection recover its vertical living cells, and GraphSAGE classifies room types from topology alone.
A comfort-analysis copilot for architectural layouts. Models six senses as one coupled system, so you can see which senses move when you change one thing. LangGraph agent, FastAPI, React. MaCAD '26 Winner (June 2026), IAAC. Built by a team of 4, now in ongoing development as my thesis.
To associate your repository with the macad topic, visit your repo's landing page and select "manage topics."