Mastering Unstructured Data with AI and ML
summary
Shell describes concrete systems that turn documents, sensor streams and contracts into actionable assets using NLP, ML and IoT pipelines with partners like C3.ai. Claim: about 80% unstructured data and a global data sphere hitting 175 zettabytes. Built: taxonomy, data governance teams, smart authoring and predictive maintenance models for control valves that run on historical vibration and telemetry. Scale examples: one facility holds ~120,000 documents, search wastes ~55% of time, and spare-part optimization reports measurable cost savings.
AI-written summary · about
chapters (13)
- 0:00Introduction and vision
- 1:25What is unstructured data
- 3:36AI and ML use cases
- 7:12Information management importance
- 10:48Retail data challenges
- 13:24Data governance and quality
- 15:30Predictive maintenance with AI
- 18:30Shell’s unique AI approach
- 21:42Unstructured data in documents
- 24:56Interpreting sensor data
- 27:04Pattern identification in operations
- 28:55In-house vs outsourced AI
- 30:01Closing remarks
related talks
39mBig Data Information Architecture for AI
11mStrengthening the data value chain
27mMarrying Structured + Unstructured Data
46mBusiness: AI as an enabler for unlocking Industry Data: Monetizing Enterprise Value
08mLeveraging data insights to increase efficiency and innovation
25mHow AI Transforms Enterprise Data Pipelines with 85% Gains
46mRobotization of Unstructured Data Processing
51mCloud data and AI security in 2024: What you need to know