Connecting the Dots: Unleash the magic of AI in IOT
summary
Claims that production IoT anomaly detection and preventive maintenance are feasible in pure Ruby, showing a prototype that trains time‑series forecasting models, detects outliers and alerts on deviations, and reruns nightly retraining. Uses a Ruby machine‑learning stack (Numo-like math libs, Daru/Rover dataframes, a scikit‑learn analogue, FastText and plotting gems) plus Prophet RB for forecasting; models serialize in ~10 lines and flags showed three to four clear anomaly incidents in sample data. Demo plots predictions vs actuals and reports workable detection accuracy for the shown dataset.
AI-written summary · about
chapters (11)
- 0:00Introduction and curiosity
- 3:38Connecting IoT AI and Ruby
- 5:01Time series data in IoT
- 6:04Challenges in IoT systems
- 7:50Machine learning with Ruby
- 9:15Ruby libraries for AI
- 11:40Anomaly detection in IoT
- 13:49Forecasting for abnormal usage
- 17:39Demo and results
- 18:42Why choose Ruby for ML
- 20:28Q&A session
related talks
20mAutomating Data Science for the IOT
21mForecasting the Future: An Introduction to Machine Learning for...
37mFrom Sensors to Insights at the Edge: Applying AI to IoT Data
35mIntelligent Ruby: Getting Started with Machine Learning
51mMachine Learning with Ruby
52mApplying AI to IoT Data
16mFrom OT Data Challenges to AI-Driven Operations on
01h 09mIoT Realized - The Connected Car