Machine Learning Terminology and Process
AWS Skill Builder
Last updated June 5, 2024
Machine Learning Terminology and Process is a free, one-hour course that walks through the full journey a machine learning project takes, one step at a time. It follows the data through nine stages: framing the business problem, turning it into a machine learning problem, collecting and integrating a dataset, preparing and cleaning that data, visualizing and analyzing it, engineering features, training the model, evaluating it, and finally checking the result against the original business goal. Along the way it introduces the working vocabulary of ML and explains it where it appears — terms like overfitting and underfitting, the bias-variance trade-off, cross-validation, and the precision-recall measures used to judge a model. The treatment is more technical than a decision-maker primer: it is built for people who want to understand what actually happens inside each phase, not just that the phases exist.
What you'll learn
- The nine steps of the ML process (business problem → business-goal evaluation)
- The working vocabulary of ML, explained in context
- Developing, cleaning, and preparing a dataset
- What feature engineering is, and common techniques
- Reading model evaluation: over/underfitting, bias-variance, precision/recall
- Checking a model against its business goal
Frequently asked questions about Machine Learning Terminology and Process
Who is Machine Learning Terminology and Process for?
Developers and technically minded learners who want to understand what happens inside each phase of a machine learning project, with the terminology that goes with it.
Is Machine Learning Terminology and Process free?
Yes — Machine Learning Terminology and Process is completely free to take.
What are the prerequisites for Machine Learning Terminology and Process?
None stated; some prior exposure to machine learning concepts helps.
Why we suggest this course
A detailed, step-by-step grounding in how a machine learning project really runs — from a business problem all the way to evaluating the result — with the field's core terminology explained in the context where it's used. A good fit for someone who has met ML at a high level and now wants the next layer of detail. One thing to know: it pitches higher than an introductory primer, assuming you're ready for real terms like the bias-variance trade-off and precision-recall.