Free Texts to Knowledge Graphs - Holy Grail of Semantic Computing on an Industrial Scale
summary
Claims a working pipeline that turns PDFs into JSON with layout, extracts n‑grams, builds word‑level graphs and maps them into active ontologies implemented in F‑logic 2, then runs declarative gap queries to surface semantic effects across clauses. Applied at FunderNation GmbH for contract versioning, the hybrid symbolic+statistical method flags non‑obvious impacts (example: a change in paragraph 21.2 adding Venture Capital investments alters remuneration in 19.1) and replaces much manual analyst work (analysts typically find ~65% of Differences).
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:43Free text to knowledge graphs
- 2:31Semaphora and its tech
- 3:36Active ontologies
- 5:21Ontologies and inferencing
- 6:45New knowledge examples
- 7:47Automating text extraction
- 9:34Regulatory documents
- 11:00Natural language understanding
- 12:25Investment business use case
- 14:13Hybrid approach
- 17:27Summary
related talks
39mAdventures in Understanding Documents
37mA novel way to manage and represent unstructured clinical data
37mCorinna Cortes, Head of Research, Google
30mFast and scalable domain-specific knowledge graphs generation
38mPanel Discussion: Creating an Open Source Data...
46mData Engineering for Graph-based Retrieval Augmented Generation GraphRAG using Neo4j
32mDataEngBytes 2023 - PER-T1-11 - Mark Kitson
30mHow Graph Data Science Can Turbocharge Your Knowledge Graph