Take Control of the Data of You
summary
Personal health system collects four years of wearables, Bluetooth scales, Untappd beer logs and runs hourly Azure Functions to store data in Cosmos DB, score it with Azure Machine Learning models and surface results through the Microsoft Bot Framework, LUIS and Cortana. The classifier predicts viral onset with ~92% confidence two days early, flagged stress patterns tied to travel, and helped hit a 4,042 km walking/running goal; code and deployment scripts are published on GitHub.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 0:37What is Personalized Medicine
- 1:12Rosalind Franklin
- 1:40Ricardo Zarate
- 2:27What is Human
- 5:07Scott Hanselman
- 15:04Artificial Intelligence
- 15:37Architecture
- 17:36Azure Functions
- 21:38Azure Machine Learning
- 28:31Two Class Classification
- 29:53Split Your Data
- 31:43Evaluate Your Models
- 35:44Power BI
- 40:41QA Dashboard
- 43:23Bot Framework
- 47:40Language Understanding
- 50:35Testing
- 55:54Publishing
related talks
55mTake Control of the Data of You
01h 00mTake Control of the Data of You
44mQuantified Self: Analyzing the Big Data of our Daily Life
14mDoes fitness data make the average person healthier?
19mEnterprise AI-Powered Predictive Maintenance Platform
01h 00mIntelligent Interfaces with Cognitive APIs, BOTs and IOT
55mSocial Listening for Richer Data, Unstructured Analytics
27mLive and Machine Learn