Introduction to machine learning

Business owner, manager, employee or student starting with no code — I can say what a machine can learn from my data, and what it cannot.

Level 1 — no code

You keep hearing that machine learning could help your business or your department, and you want to judge that claim yourself before spending money on it. You do not write code and you do not need to.

What you'll be able to do

• Explain in plain words how a model learns from examples rather than from rules
• Tell a prediction problem apart from a grouping problem and from a task no model should do
• Look at your own records and say whether they could train anything useful
• Train a small classifier yourself with Teachable Machine, and read its mistakes
• Read accuracy honestly, spot a biased dataset and refuse a bad model
• Write a one-page brief a developer or a vendor can actually work from

Modules

• ML1 — What learning from data really means — Rules versus examples, and the vocabulary you need to follow any later conversation.
• ML2 — The three kinds of problem — Prediction, grouping, and the category no model should touch.
• ML3 — Your own data — Look honestly at the records you already keep and judge whether they can train anything.
• ML4 — Train one yourself — A real classifier, trained in the browser, with its mistakes read out loud.
• ML5 — Judging a model honestly — Accuracy, bias and the questions to ask a vendor before you pay.
• ML6 — Deciding and briefing — Turn all of it into one decision and one page.

Capstone project

One honest machine-learning brief — Take one real problem from your own business, department or school. Describe the data you actually hold, train a demonstration classifier in Teachable Machine, read its errors, then write the one-page brief that says whether a model is worth building — including the answer "no, use a rule".

What you earn

• A trained demo model and a written reading of its errors
• A one-page machine-learning brief for your own organisation
• LES Institute certificate — Introduction to Machine Learning

Requirements

• Reading and writing in French or English
• A phone or laptop with a browser
• Some records of your own work: sales, stock, attendance, requests