Big Data Analytics, Machine Learning, Real Time Processing
summary
Shows integrating analytic models built with R, Spark MLlib, H2O into a live streaming pipeline on Hadoop and reusing the exact binary model in production. Builds a predictive maintenance flow that scores each sensor event and flags parts to scrap, achieving about 3.6–3.9 ms end-to-end with ~0.3 ms per model application. Demonstrates k‑means and GLM examples, details data prep/validation steps, and proves the stream processing deployment works at required throughput.
AI-written summary · about
chapters (7)
- 0:00Start
- 0:53Analyze and Act on Critical Business Moments
- 3:37Real World Examples of Machine Learning
- 11:35Exploratory Data Analysis
- 15:08Model Improvement
- 18:28Advanced Analytics and Big Data Tools for Data Scientists
- 30:48Fast Data Architecture for Predictive Maintenance
related talks
38mRealtime Data Activation - Analytics, Intelligence & Decision Making
50mScalable Data Science & Deep Learning with H2O
30mSmart Streams: A Real-time framework for scoring Big and Fast Data
40mHow to leverage the Apache Kafka Ecosystem to productionize Machine Learning
11mScaling R with Spark
19mBig Data
01h 26mData Driven Action: a primer on Data Science
40mReal-time Learning with Spark