Machine Learning models utilization
summary
BlackSwan Technologies builds a web app called pet Miner that converts free text into Knowledge Graphs using NER, relationship extraction and a deterministic coreference resolution step, with custom entity and relation types for politically exposed persons. The demo shows automatic graph visualizations (nodes for people, titles, orgs), works end-to-end but has recurring errors (mismerged entities, wrong residence via parent link). Planned next steps are entity linking across documents and a lightweight classifier to limit expensive model runs.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:23What is a Knowledge Graph
- 5:26Name Entity Recognition Model
- 7:25Relationship Extraction Model
- 9:50Reference Resolution Model
- 13:30Knowledge Graphs
- 17:10Solution
- 19:10Demo
- 22:02Graphing
- 28:33Summary
related talks
35mIndustrial Applications of Machine Learning in Search
46mBlack Magic: How to apply Machine Learning to real-world problems
01h 30mBecome a data-driven Organization with Machine Learning
34mExposing Dark Data in the enterprise with custom NLP
21mRelationships Matter: Mining relationships using Deep Learning
57mMachine Learning with JavaScript
24mEmpowering Lifelong Learning: Transforming Data into Insights With GraphRAG
42mBeyond Vectors: Evolving GenAI through Transformative Tools and Methods