From Sensors to Insights at the Edge: Applying AI to IoT Data

on YouTube ↗37m

" summary

A pipeline reads sensor data from an STM32MP1 over MQTT into Node-RED, stores timestamped records in Cloudant, trains a logistic regression model in Watson Studio using Spark and a Jupyter notebook on roughly 100 labeled samples (≈50 normal, 50 hot), then pushes the model coefficients back to the device so predictions and alerts run locally at ~3s sampling; dashboards visualize temperature, humidity and IMU traces and show real‑time alerting.

AI-written summary · about

" chapters (8)

1. 0:00Start
2. 0:58Live Demo
3. 12:32Node Red
4. 13:59The Ibm Cloud
5. 16:39Node-Red in the Cloud
6. 27:01Watson Studio
7. 28:32Spark Session
8. 36:29Conclusion

" related talks

• 52mApplying AI to IoT DataJohn Walicki · Devnexus 2020
• 45mIoT End to EndJohn Walicki · Global IoT DevFest 2017
• 35mData science and Internet of Things on the edgeMario Bonamigo · PyData London 2019
• 14mUnlock the power of IoT: Intro to NTT’s edge analytics platformBill Baver · AWS re:Invent 2023
• 32mExtending machine learning to industrial IoT applications at the edge with AWSIan Perez Ponce · Arm AIoT Dev Summit 2019
• 20mAutomating Data Science for the IOTRuban Phukan · Data By the Bay 2016
• 37mPracticing Data Science in the cloudDr. Robert Hoffmann · WeAreDevelopers AI Congress Vienna 2018
• 59mML and the IoT: Living on the EdgeBrandon D Satrom · NDC Oslo 2019