Decoding Regulations Using Machine Learning
summary
Regulatory text piles up — 81,000 pages since 2015 — and comes as messy PDFs, tables and flowcharts; IBM Emilia is pre-trained on federal, state and local regs and ingests those documents to extract regulatory terms, maps regulatory terms to business terms, and ranks which business items matter most (example: flavored milk). Emilia uses recurrent neural nets, deep learning and NLP to turn unstructured rules into a 360-degree view of cost and revenue impact and even flags new product opportunities like 1% flavored milk.
AI-written summary · about
related talks
46mMachine Learning and Digital Regulation
23mSynthetic data that meets enterprise compliance needs
29mEY Aleph: deep learning applied to jurimetrics practice
15mMachine learning: Necessity vs obligation
44mContent Classification: NLP vs. AI in the Age of Data Vectorization
44mExtracting Structured Data from Legal Documents
24mMachine Intelligence gets real
23mChallenges in using machine learning for financial monitoring