Predict the future (or don't) with machine learning in applications
summary
Demonstrates building in-database models with Oracle Machine Learning (AutoML UI, OML4Py, SQL) to train a Random Forest and SVM (Gaussian) on a 60/40 split, achieving ~75% balanced accuracy for the top model; shows predictor importance with age, years_residence, education as top features, dynamic scoring of single records and millions of rows with prediction probabilities, selecting customers >80% likely to buy, and deploying models to OML Services (REST endpoints) for production use.
AI-written summary · about
chapters (6)
related talks
31mIn-database machine learning: Automate, scale, deploy
26mOracle Machine Learning: what's new and what's next?
31mAppDev with Machine Learning and AutoML
01h 39mApplying Rigorous Machine Learning Methods in Business Strategy
28mAutoML—hype vs. value
56mPredictive Analytics for Practitioners
37mForecasting using machine learning methods
59mPredicting the future as a service with Azure ML and R