Frame the Problem Before Choosing a Model
Learners can decide whether a machine-learning model is a suitable response to a practical problem before they choose an algorithm.
Executable courseRead the lesson pages, run the browser practice notebook, and keep the downloadable fallback pack if you want a local copy.
Module sequence
- Why a Model Is Not the First QuestionUnit 1 · 15-20 minutes
- Prediction, Description, Automation, and Decision SupportUnit 2 · 20-30 minutes
- Target, Unit, Time Horizon, and ActionUnit 3 · 25-35 minutes
- Users, Affected People, and Failure ModesUnit 4 · 25-35 minutes
- Baselines, Rules, Statistics, and the No-Model OptionUnit 5 · 25-35 minutes
- Feasibility: Data, Labels, Timing, and CostUnit 6 · 25-35 minutes
- Project Step: Problem Brief and Harm StatementUnit 7 · 30-45 minutes
- Activity: Write a Machine-Learning Problem BriefPractice activity · 35-50 minutes
- Module 1 Assessment: Problem Framing CheckModule check · 20-30 minutes