Lesson 3: Machine Learning: Learning from Data

Learning Objectives Define Machine Learning (ML) and explain how it differs from traditional programming. Identify the three main types of machine learning: supervised, unsupervised, and reinforcement. Explain the critical role of data in training a machine learning model. Describe the basic workflow of an ML project: data collection, training, and prediction. Provide at least three real-world examples of machine learning they use every day. Recognize the importance of using large, unbiased datasets for training. Explain what a feature and a label are in the context of a dataset. Ever wonder how TikTok or YouTube knows exactly what video you want to watch next? 🤖 It's not magic, it's machine learning! In this lesson, we'll explore how we can teach computer...

TermDefinitionExample Machine Learning (ML)A type of artificial intelligence that gives computers the ability to learn from data and find patterns on their own, without being explicitly programmed for every single step.Instead of writing code with `if` statements for every type of spam email, we 'train' a model by showing it thousands of examples of spam and not-spam emails, and it learns how to tell the difference itself. DataThe information, such as images, text, or numbers, that is used to teach a machine learning model.For a weather prediction model, the data would be past information about temperature, humidity, wind speed, and whether it rained. TrainingThe process of feeding data to a machine learning model so it can learn to find patterns or make correct decisions.Showin...

The Basic ML Workflow 1. Gather Data -> 2. Train Model -> 3. Make Predictions This is the fundamental three-step pattern for almost any machine learning project. You need information to learn from, a process for learning, and a way to use what was learned. The GIGO Principle Garbage In, Garbage Out A machine learning model is only as good as the data it's trained on. If you train a model with bad, incomplete, or biased data, it will make bad, incomplete, or biased predictions. The Three Learning Styles Supervised (labeled data), Unsupervised (unlabeled data), Reinforcement (rewards/penalties) These are the main ways a machine can learn. We choose a style based on the problem we want to solve and the kind of data we have available.