Machine Learning Basics: How Algorithms Learn from Data

Understanding machine learning basics is the foundation of this guide. Below we break it down step by step so you can act with confidence.

Machine learning is the engine behind modern AI. Instead of writing rules for every possible scenario, you give the algorithm data and let it discover the patterns itself. The result is a model that can make predictions or decisions on new data it has never seen before.

Understanding the basics of machine learning — the types, the workflow, and the evaluation methods — is essential for everything that follows in this course.

Machine Learning Basics: Three types of machine learning

Machine learning algorithms fall into three main categories:

Supervised learning — the algorithm learns from labelled examples. You provide input-output pairs (e.g. house features → price), and the model learns the mapping. Used for classification (spam/not-spam) and regression (predicting a number).
Unsupervised learning — the algorithm finds patterns in unlabelled data. No correct answers are provided. Used for clustering (grouping similar customers), dimensionality reduction, and anomaly detection.
Reinforcement learning — the algorithm learns by taking actions in an environment and receiving rewards or penalties. Used in game playing, robotics, and recommendation systems.

Most real-world ML applications use supervised learning. This course covers supervised and unsupervised learning in depth, with a foundation in reinforcement learning concepts.

The machine learning workflow

Every ML project follows a similar pipeline:

Define the problem — what are you predicting or discovering?
Collect and clean data — garbage in, garbage out. Data quality determines model quality.
Explore and visualise — understand distributions, correlations, and outliers.
Feature engineering — create or transform variables to help the model learn.
Select and train a model — choose an algorithm, feed it training data.
Evaluate — measure performance on data the model has never seen (test set).
Deploy and monitor — put the model into production and track its performance over time.

The most time-consuming step is usually data preparation — professional data scientists spend 60–80% of their time cleaning and preparing data, not building models.

How to evaluate machine learning models

A model is only useful if it generalises — performs well on new, unseen data. Key evaluation concepts:

Training set / test set split — train on 80% of data, evaluate on 20%
Accuracy — percentage of correct predictions (misleading with imbalanced classes)
Precision and recall — how many positive predictions are correct, and how many actual positives are found
F1 score — harmonic mean of precision and recall, balances both
Overfitting — the model memorises training data but fails on new data (high training accuracy, low test accuracy)
Underfitting — the model is too simple to capture the patterns (low accuracy on both)

Cross-validation (splitting data into multiple folds and averaging results) gives more reliable evaluation than a single train/test split.

When evaluating machine learning basics, start with the fundamentals: what problem it solves, who it serves, and how it compares to the alternatives buyers consider first.

Start with the fundamentals. Define who you are serving, the single outcome you want, and the budget you are comfortable committing before you compare any options.

A practical rule is to favour clarity over cleverness. The choice that is easiest to understand, spell, remember and act on tends to win over the long run.

Context changes the answer. What is optimal for a fast-moving startup is rarely optimal for an established brand, so weigh your own stage, market and ambitions.

Avoid the common mistakes: rushing the decision, copying competitors blindly, and ignoring the total cost of ownership beyond the very first year of use.

Think in terms of trade-offs rather than a single perfect answer. Every option trades memorability, cost, reach and flexibility in a slightly different mix.

Document your reasoning. Writing down why you chose one path makes it far easier to revisit the decision later and to explain it to partners and investors.

Test with real people. A quick check with a handful of your actual customers will surface problems that no amount of internal debate ever reveals.

Plan for growth. The option that fits today should still make sense when your audience, catalogue or team is several times larger than it is right now.

Finally, revisit the decision periodically. Markets, tools and best practices move quickly, and a choice that was ideal last year may deserve a fresh look today.