Open-source toolkit to extract structured knowledge graphs from documents and tables — power analytics, digital twins, and AI-driven assistants.

CauseNet: Towards a Causality Graph Extracted from the Web

Extraction of graphs from skeletonised images.

Programmatic data access layer for LLM agents - navigate context, don't memorize it

Generates training data for graph extraction.

Hateful Meme challenge, Knowledge graph based approach

GUI for visualising graph extraction on skeletonised images.

Network diagram to structured data converter for Visio draw.io and photos

mask2graph converts 2D/3D binary masks into deterministic geometric graphs (nodes, edges, radii, and metadata) for downstream analysis, simulation, and learning systems.

Binary release of CloudSeg and CloudGraph for TomatoPGT digital twin modeling

Processing of bibliographic files

A complete end-to-end pipeline for extracting structured data from chart and graph images

Demonstrates object graph extraction from a relational database using plain JDBC (JdbcClient) — flat SQL JOIN results assembled into a three-level hierarchy without ORM or pagination.

CauseNet: Towards a Causality Graph Extracted from the Web

Code and lightweight audit artifacts for auditable repair of scientific reasoning graph extraction on a 350-row benchmark.

Machine Learning for Medical Image Processing. Project done for Charité, the university hospital of Berlin. The aim was to segment coronary arteries and then extract a graph from it, in order to aid detection of coronary artery disease (CAD).

Demonstrates object graph extraction using JPA entities and Blaze Persistence Entity Views — graph fetching without N+1, without manual mapping, without pagination.

Demonstrates object graph extraction using jOOQ MULTISET — nested SQL collections mapped directly into a three-level hierarchy without ORM, projections, or pagination.